Oracle Database

JDBC Developers Guide and Reference, 11g Release 1 (11.1)

B31224-03

September 2007 This book describes how to use Oracle JDBC drivers to develop powerful Java database applications.

Oracle Database JDBC Developer's Guide and Reference, 11g Release 1 (11.1) B31224-03 Copyright 1999, 2007, Oracle. All rights reserved. Primary Author: Pfaeffle Tulika Das, Venkatasubramaniam Iyer, Elizabeth Hanes Perry, Brian Wright, Thomas Brian Martin

Contributing Author:

Contributor: Kuassi Mensah, Douglas Surber, Paul Lo, Ed Shirk, Tong Zhou, Jean de Lavarene, Rajkumar Irudayaraj, Ashok Shivarudraiah, Angela Barone, Rosie Chen, Sunil Kunisetty, Joyce Yang, Mehul Bastawala, Luxi Chidambaran, Srinath Krishnaswamy, Longxing Deng, Magdi Morsi, Ron Peterson, Ekkehard Rohwedder, Catherine Wong, Scott Urman, Jerry Schwarz, Steve Ding, Soulaiman Htite, Anthony Lai, Prabha Krishna, Ellen Siegal, Susan Kraft, Sheryl Maring The Programs (which include both the software and documentation) contain proprietary information; they are provided under a license agreement containing restrictions on use and disclosure and are also protected by copyright, patent, and other intellectual and industrial property laws. Reverse engineering, disassembly, or decompilation of the Programs, except to the extent required to obtain interoperability with other independently created software or as specified by law, is prohibited. The information contained in this document is subject to change without notice. If you find any problems in the documentation, please report them to us in writing. This document is not warranted to be error-free. Except as may be expressly permitted in your license agreement for these Programs, no part of these Programs may be reproduced or transmitted in any form or by any means, electronic or mechanical, for any purpose. If the Programs are delivered to the United States Government or anyone licensing or using the Programs on behalf of the United States Government, the following notice is applicable: U.S. GOVERNMENT RIGHTS Programs, software, databases, and related documentation and technical data delivered to U.S. Government customers are "commercial computer software" or "commercial technical data" pursuant to the applicable Federal Acquisition Regulation and agency-specific supplemental regulations. As such, use, duplication, disclosure, modification, and adaptation of the Programs, including documentation and technical data, shall be subject to the licensing restrictions set forth in the applicable Oracle license agreement, and, to the extent applicable, the additional rights set forth in FAR 52.227-19, Commercial Computer Software--Restricted Rights (June 1987). Oracle USA, Inc., 500 Oracle Parkway, Redwood City, CA 94065. The Programs are not intended for use in any nuclear, aviation, mass transit, medical, or other inherently dangerous applications. It shall be the licensee's responsibility to take all appropriate fail-safe, backup, redundancy and other measures to ensure the safe use of such applications if the Programs are used for such purposes, and we disclaim liability for any damages caused by such use of the Programs. Oracle, JD Edwards, PeopleSoft, and Siebel are registered trademarks of Oracle Corporation and/or its affiliates. Other names may be trademarks of their respective owners. The Programs may provide links to Web sites and access to content, products, and services from third parties. Oracle is not responsible for the availability of, or any content provided on, third-party Web sites. You bear all risks associated with the use of such content. If you choose to purchase any products or services from a third party, the relationship is directly between you and the third party. Oracle is not responsible for: (a) the quality of third-party products or services; or (b) fulfilling any of the terms of the agreement with the third party, including delivery of products or services and warranty obligations related to purchased products or services. Oracle is not responsible for any loss or damage of any sort that you may incur from dealing with any third party.

Contents
Preface ............................................................................................................................................................... xix
Audience..................................................................................................................................................... xix Documentation Accessibility ................................................................................................................... xix Related Documents ................................................................................................................................... xx Conventions .............................................................................................................................................. xxii

Whats New.................................................................................................................................................... xxiii

New Features for Release 1 (11.1) ......................................................................................................... Desupported Features ............................................................................................................................ Interface Changes .................................................................................................................................... xxiii xxiv xxiv

Part I 1

Overview

Introducing JDBC
Overview of Oracle JDBC Drivers ........................................................................................................ Common Features of Oracle JDBC Drivers .................................................................................... Choosing the Appropriate Driver.................................................................................................... Feature Differences Between JDBC OCI and Thin Drivers.......................................................... Environments and Support .................................................................................................................... Supported JDK and JDBC Versions................................................................................................. JNI and Java Environments .............................................................................................................. JDBC and IDEs.................................................................................................................................... Feature List ................................................................................................................................................ 1-1 1-2 1-4 1-4 1-4 1-5 1-5 1-5 1-5

Getting Started
Version Compatibility for Oracle JDBC Drivers................................................................................ Verification of a JDBC Client Installation ........................................................................................... Check the Installed Directories and Files ....................................................................................... Check the Environment Variables ................................................................................................... Ensure that the Java Code Can Be Compiled and Run ................................................................ Determine the Version of the JDBC Driver .................................................................................... Test JDBC and the Database Connection........................................................................................ Basic Steps in JDBC ................................................................................................................................. Importing Packages ........................................................................................................................... 2-1 2-2 2-2 2-3 2-4 2-5 2-5 2-7 2-8

iii

Opening a Connection to a Database .............................................................................................. 2-8 Creating a Statement Object ............................................................................................................. 2-9 Running a Query and Retrieving a ResultSet Object.................................................................... 2-9 Processing the ResultSet Object .................................................................................................... 2-10 Closing the ResultSet and Statement Objects.............................................................................. 2-10 Making Changes to the Database ................................................................................................. 2-11 Committing Changes...................................................................................................................... 2-11 Closing the Connection .................................................................................................................. 2-12 Sample: Connecting, Querying, and Processing the Results ........................................................ 2-12 Stored Procedure Calls in JDBC Programs....................................................................................... 2-13 PL/SQL Stored Procedures ........................................................................................................... 2-13 Java Stored Procedures................................................................................................................... 2-14 Processing SQL Exceptions ................................................................................................................. 2-14

Part II 3

Oracle JDBC

JDBC Standards Support

Support for JDBC 2.0 Standard.............................................................................................................. Data Type Support ............................................................................................................................. Standard Feature Support................................................................................................................. Extended Feature Support ................................................................................................................ Standard versus Oracle Performance Enhancement APIs ........................................................... Support for JDBC 3.0 Standard.............................................................................................................. Transaction Savepoints...................................................................................................................... Creating a Savepoint .................................................................................................................. Rolling Back to a Savepoint ....................................................................................................... Releasing a Savepoint................................................................................................................. Checking Savepoint Support..................................................................................................... Savepoint Notes .......................................................................................................................... Retrieval of Auto-Generated Keys................................................................................................... java.sql.Statement ....................................................................................................................... Sample Code ................................................................................................................................ Limitations ................................................................................................................................... JDBC 3.0 LOB Interface Methods..................................................................................................... Result Set Holdability........................................................................................................................ Support for JDBC 4.0 Standard.............................................................................................................. 3-1 3-2 3-2 3-2 3-2 3-2 3-3 3-3 3-3 3-4 3-4 3-4 3-4 3-4 3-5 3-5 3-5 3-6 3-6

Oracle Extensions
Overview of Oracle Extensions ............................................................................................................. Features of the Oracle Extensions ........................................................................................................ Database Management Using JDBC................................................................................................ Support for Oracle Data Types......................................................................................................... Support for Oracle Objects................................................................................................................ Support for Schema Naming ............................................................................................................ DML Returning .................................................................................................................................. Accessing PL/SQL Index-by Tables................................................................................................ 4-1 4-2 4-2 4-2 4-3 4-4 4-4 4-4

iv

Oracle JDBC Packages ............................................................................................................................. 4-5 Package oracle.sql .............................................................................................................................. 4-5 Package oracle.jdbc ......................................................................................................................... 4-10 Oracle Character Data Types Support ............................................................................................... 4-10 SQL CHAR Data Types .................................................................................................................. 4-10 SQL NCHAR Data Types............................................................................................................... 4-10 Class oracle.sql.CHAR.................................................................................................................... 4-11 Additional Oracle Type Extensions ................................................................................................... 4-13 Oracle ROWID Type....................................................................................................................... 4-13 Oracle REF CURSOR Type Category ........................................................................................... 4-14 Oracle BINARY_FLOAT and BINARY_DOUBLE Types ......................................................... 4-16 Oracle SYS.ANYTYPE and SYS.ANYDATA Types ................................................................... 4-16 The oracle.jdbc Package ................................................................................................................. 4-19 Interface oracle.jdbc.OracleConnection ................................................................................ 4-21 Interface oracle.jdbc.OracleStatement................................................................................... 4-22 Interface oracle.jdbc.OraclePreparedStatement .................................................................. 4-22 Interface oracle.jdbc.OracleCallableStatement .................................................................... 4-22 Interface oracle.jdbc.OracleResultSet .................................................................................... 4-23 Interface oracle.jdbc.OracleResultSetMetaData................................................................... 4-23 Class oracle.jdbc.OracleTypes................................................................................................ 4-23 Method getJavaSqlConnection............................................................................................... 4-25 DML Returning ..................................................................................................................................... 4-26 Oracle-Specific APIs........................................................................................................................ 4-26 Running DML Returning Statements........................................................................................... 4-27 Example of DML Returning .......................................................................................................... 4-27 Limitations of DML Returning...................................................................................................... 4-28 Accessing PL/SQL Index-by Tables ................................................................................................... 4-28 Overview .......................................................................................................................................... 4-29 Binding IN Parameters ................................................................................................................... 4-29 Receiving OUT Parameters............................................................................................................ 4-31 Type Mappings................................................................................................................................ 4-32

Features Specific to JDBC Thin

Overview of JDBC Thin Client ............................................................................................................. Additional Features Supported ............................................................................................................. Support for Applets ........................................................................................................................... Default Support for Native XA ........................................................................................................ JDBC in Applets ....................................................................................................................................... Connecting to the Database Through the Applet.......................................................................... Connecting to a Database on a Different Host Than the Web Server ........................................ Using the Oracle Connection Manager.................................................................................... Using Signed Applets................................................................................................................. Using Applets with Firewalls........................................................................................................... Configuring a Firewall for Applets that use the JDBC Thin Driver .................................... Writing a URL to Connect Through a Firewall ...................................................................... Packaging Applets ............................................................................................................................. Specifying an Applet in an HTML Page ......................................................................................... 5-1 5-1 5-2 5-2 5-2 5-2 5-3 5-4 5-5 5-6 5-6 5-7 5-8 5-8

CODE, HEIGHT, and WIDTH .................................................................................................. 5-8 CODEBASE.................................................................................................................................. 5-9 ARCHIVE..................................................................................................................................... 5-9

Features Specific to JDBC OCI Driver

OCI Connection Pooling ......................................................................................................................... 6-1 Transparent Application Failover ......................................................................................................... 6-1 OCI Native XA .......................................................................................................................................... 6-1 OCI Instant Client .................................................................................................................................... 6-1 Overview of Instant Client................................................................................................................ 6-2 Benefits of Instant Client ................................................................................................................... 6-2 JDBC OCI Instant Client Installation Process ................................................................................ 6-3 Usage of Instant Client ...................................................................................................................... 6-4 Patching Instant Client Shared Libraries ........................................................................................ 6-4 Regeneration of Data Shared Library and ZIP files ...................................................................... 6-5 Database Connection Names for OCI Instant Client .................................................................... 6-5 Environment Variables for OCI Instant Client .............................................................................. 6-7 Instant Client Light (English) ................................................................................................................ 6-8 Globalization Settings........................................................................................................................ 6-9 Operation............................................................................................................................................. 6-9 Installation........................................................................................................................................ 6-10

Server-Side Internal Driver

Overview of the Server-Side Internal Driver ..................................................................................... Connecting to the Database.................................................................................................................... Exception-Handling Extensions ............................................................................................................ Session and Transaction Context ........................................................................................................... Testing JDBC on the Server .................................................................................................................... Loading an Application into the Server ............................................................................................... 7-1 7-1 7-3 7-4 7-4 7-5

Part III 8

Connection and Security

Data Sources and URLs

Data Sources .............................................................................................................................................. 8-1 Overview of Oracle Data Source Support for JNDI ...................................................................... 8-1 Features and Properties of Data Sources ........................................................................................ 8-2 Creating a Data Source Instance and Connecting ......................................................................... 8-6 Creating a Data Source Instance, Registering with JNDI, and Connecting............................... 8-6 Supported Connection Properties ................................................................................................... 8-8 Using Roles for SYS Login ................................................................................................................ 8-8 Configuring Database Remote Login.............................................................................................. 8-8 Bequeath Connection and SYS Logon............................................................................................. 8-9 Properties for Oracle Performance Extensions ........................................................................... 8-10 Database URLs and Database Specifiers .......................................................................................... 8-11

vi

JDBC Client-Side Security Features

Support for Oracle Advanced Security ................................................................................................ 9-1 Support for Login Authentication ........................................................................................................ 9-3 Support for Strong Authentication ....................................................................................................... 9-3 Support for OS Authentication ............................................................................................................. 9-4 Configuration Steps for Linux.......................................................................................................... 9-4 Configuration Steps for Windows ................................................................................................... 9-5 JDBC Code Using OS Authentication ............................................................................................. 9-6 Support for Data Encryption and Integrity ......................................................................................... 9-6 JDBC OCI Driver Support for Encryption and Integrity.............................................................. 9-7 JDBC Thin Driver Support for Encryption and Integrity............................................................. 9-8 Setting Encryption and Integrity Parameters in Java ................................................................... 9-9 Support for SSL ..................................................................................................................................... 9-11 Managing Certificates and Wallets............................................................................................... 9-12 Keys and certificates containers .................................................................................................... 9-12 Support for Kerberos ............................................................................................................................ 9-13 Configuring Oracle Database to Use Kerberos ........................................................................... 9-13 Code Example.................................................................................................................................. 9-14 Support for RADIUS ............................................................................................................................ 9-18 Configuring Oracle Database to Use RADIUS ........................................................................... 9-18 Code Example.................................................................................................................................. 9-19 Secure External Password Store ......................................................................................................... 9-21

10

Proxy Authentication
About Proxy Authentication ............................................................................................................... Types of Proxy Connections ................................................................................................................ Creating Proxy Connections ................................................................................................................ Closing a Proxy Session ....................................................................................................................... Caching Proxy Connections ................................................................................................................ 10-1 10-2 10-3 10-4 10-5

Part IV 11

Data Access and Manipulation

Accessing and Manipulating Oracle Data

Data Type Mappings............................................................................................................................. Table of Mappings .......................................................................................................................... Notes Regarding Mappings........................................................................................................... Data Conversion Considerations ....................................................................................................... Standard Types Versus Oracle Types .......................................................................................... Converting SQL NULL Data ......................................................................................................... Testing for NULLs .......................................................................................................................... Result Set and Statement Extensions ................................................................................................ Comparison of Oracle get and set Methods to Standard JDBC ................................................... Standard getObject Method........................................................................................................... Oracle getOracleObject Method.................................................................................................... Summary of getObject and getOracleObject Return Types ...................................................... Other getXXX Methods .................................................................................................................. 11-1 11-1 11-3 11-3 11-4 11-4 11-4 11-5 11-5 11-6 11-6 11-7 11-9

vii

Return Types of getXXX Methods ......................................................................................... Special Notes about getXXX Methods ................................................................................ Data Types For Returned Objects from getObject and getXXX ............................................. The setObject and setOracleObject Methods............................................................................. Other setXXX Methods................................................................................................................. Input Data Binding ................................................................................................................ Method setFixedCHAR for Binding CHAR Data into WHERE Clauses....................... Using Result Set Metadata Extensions............................................................................................ Using SQL CALL and CALL INTO Statements ............................................................................

11-9 11-10 11-10 11-11 11-11 11-12 11-13 11-14 11-15

12

Java Streams in JDBC

Overview ................................................................................................................................................. Streaming LONG or LONG RAW Columns .................................................................................... LONG RAW Data Conversions .................................................................................................... LONG Data Conversions ............................................................................................................... Streaming Example for LONG RAW Data.................................................................................. Avoiding Streaming for LONG or LONG RAW ........................................................................ Streaming CHAR, VARCHAR, or RAW Columns .......................................................................... Streaming LOBs and External Files ................................................................................................... Data Streaming and Multiple Columns............................................................................................ Closing a Stream .................................................................................................................................... Notes and Precautions on Streams ..................................................................................................... Streaming Data Precautions .......................................................................................................... Using Streams to Avoid Limits on setBytes and setString........................................................ Streaming and Row Prefetching ................................................................................................... 12-1 12-2 12-2 12-2 12-3 12-5 12-5 12-6 12-7 12-8 12-8 12-8 12-9 12-9

13

Working with Oracle Object Types

Mapping Oracle Objects ...................................................................................................................... Using the Default STRUCT Class for Oracle Objects ................................................................... STRUCT Class Functionality ......................................................................................................... Retrieving STRUCT Objects and Attributes................................................................................ Creating STRUCT Objects and Descriptors................................................................................. Binding STRUCT Objects into Statements................................................................................... STRUCT Automatic Attribute Buffering ..................................................................................... Creating and Using Custom Object Classes for Oracle Objects .................................................. Relative Advantages of ORAData versus SQLData................................................................... Understanding Type Maps for SQLData Implementations...................................................... Creating Type Map and Defining Mappings for a SQLData Implementation ...................... Adding Entries to an Existing Type Map ............................................................................. Creating a New Type Map ................................................................................................... Materializing Object Types not Specified in the Type Map ............................................ Understanding the SQLData Interface....................................................................................... Reading and Writing Data with a SQLData Implementation ................................................ Understanding the ORAData Interface ..................................................................................... Reading and Writing Data with a ORAData Implementation ............................................... Additional Uses for ORAData..................................................................................................... The Deprecated CustomDatum Interface.................................................................................. 13-1 13-2 13-2 13-3 13-5 13-6 13-6 13-6 13-7 13-8 13-8 13-9 13-10 13-10 13-10 13-12 13-14 13-16 13-18 13-19

viii

Object-Type Inheritance .................................................................................................................... Creating Subtypes ......................................................................................................................... Implementing Customized Classes for Subtypes..................................................................... Use of ORAData for Type Inheritance Hierarchy ............................................................. Use of SQLData for Type Inheritance Hierarchy .............................................................. JPublisher Utility.................................................................................................................... Retrieving Subtype Objects.......................................................................................................... Creating Subtype Objects............................................................................................................. Sending Subtype Objects.............................................................................................................. Accessing Subtype Data Fields ................................................................................................... Inheritance Metadata Methods ................................................................................................... Using JPublisher to Create Custom Object Classes ..................................................................... JPublisher Functionality ............................................................................................................... JPublisher Type Mappings .......................................................................................................... Describing an Object Type ................................................................................................................ Functionality for Getting Object Metadata................................................................................ Steps for Retrieving Object Metadata.........................................................................................

13-19 13-20 13-20 13-21 13-23 13-25 13-26 13-28 13-28 13-29 13-30 13-30 13-30 13-31 13-33 13-33 13-34

14

Working with LOBs and BFILEs

Oracle Extensions for LOBs and BFILEs........................................................................................... Working with BLOBs, CLOBs and NCLOBs ................................................................................... Getting and Passing BLOB, CLOB, and NCLOB Locators........................................................ Retrieving BLOB, CLOB, and NCLOB Locators ................................................................. Passing BLOB, CLOB and NCLOB Locators ....................................................................... Reading and Writing BLOB, CLOB and NCLOB Data.............................................................. Creating and Populating a BLOB, CLOB or NCLOB Column ................................................. Accessing and Manipulating BLOB, CLOB, or NCLOB Data ................................................ Data Interface for LOBs ..................................................................................................................... Working With Temporary LOBs ....................................................................................................... Using Open and Close With LOBs .................................................................................................. Working with BFILEs ......................................................................................................................... Getting and Passing BFILE Locators .......................................................................................... Reading BFILE Data...................................................................................................................... Creating and Populating a BFILE Column................................................................................ Accessing and Manipulating BFILE Data.................................................................................. Oracle SecureFiles ............................................................................................................................... 14-1 14-2 14-2 14-2 14-4 14-5 14-8 14-10 14-10 14-13 14-14 14-14 14-15 14-16 14-17 14-18 14-19

15

Using Oracle Object References

Oracle Extensions for Object References.......................................................................................... Overview of Object Reference Functionality .................................................................................. Object Reference Getter and Setter Methods .............................................................................. Key REF Class Methods ................................................................................................................. Retrieving and Passing an Object Reference ................................................................................... Retrieving an Object Reference from a Result Set ...................................................................... Retrieving an Object Reference from a Callable Statement ...................................................... Passing an Object Reference to a Prepared Statement............................................................... 15-1 15-2 15-2 15-2 15-3 15-3 15-4 15-4

ix

Accessing and Updating Object Values Through an Object Reference ..................................... 15-5 Custom Reference Classes with JPublisher ..................................................................................... 15-5

16

Working with Oracle Collections

Oracle Extensions for Collections ...................................................................................................... Choices in Materializing Collections............................................................................................ Creating Collections........................................................................................................................ Creating Multilevel Collection Types .......................................................................................... Overview of Collection Functionality ............................................................................................... ARRAY Performance Extension Methods ........................................................................................ Accessing oracle.sql.ARRAY Elements as Arrays of Java Primitive Types............................ ARRAY Automatic Element Buffering ........................................................................................ ARRAY Automatic Indexing......................................................................................................... Creating and Using Arrays .................................................................................................................. Creating ARRAY Objects ............................................................................................................... Retrieving an Array and Its Elements .......................................................................................... Retrieving the Array ............................................................................................................... Data Retrieval Methods .......................................................................................................... Comparing the Data Retrieval Methods............................................................................... Retrieving Elements of a Structured Object Array According to a Type Map ............... Retrieving a Subset of Array Elements ................................................................................. Retrieving Array Elements into an oracle.sql.Datum Array ............................................. Accessing Multilevel Collection Elements ......................................................................... Passing Arrays to Statement Objects.......................................................................................... Using a Type Map to Map Array Elements .................................................................................... Custom Collection Classes with JPublisher .................................................................................. 16-1 16-1 16-2 16-3 16-3 16-3 16-4 16-4 16-4 16-5 16-5 16-6 16-6 16-6 16-7 16-8 16-8 16-9 16-10 16-11 16-12 16-13

17

Result Set
Overview ................................................................................................................................................. Result Set Functionality and Result Set Categories Supported in JDBC 2.0........................... Oracle JDBC Implementation Overview for Result Set Enhancements .................................. Creating Scrollable or Updatable Result Sets ................................................................................. Specifying Result Set Scrollability and Updatability ................................................................. Result Set Limitations and Downgrade Rules ............................................................................ Positioning and Processing in Scrollable Result Sets .................................................................... Positioning in a Scrollable Result Set ........................................................................................... Processing a Scrollable Result Set ............................................................................................... Updating Result Sets .......................................................................................................................... Performing a DELETE Operation in a Result Set ..................................................................... Performing an UPDATE Operation in a Result Set ................................................................. Performing an INSERT Operation in a Result Set.................................................................... Avoiding Update Conflicts.......................................................................................................... Fetch Size .............................................................................................................................................. Setting the Fetch Size .................................................................................................................... Refetching Rows .................................................................................................................................. Seeing Database Changes Made Internally and Externally........................................................ Seeing Internal Changes............................................................................................................... 17-1 17-1 17-3 17-5 17-5 17-6 17-8 17-8 17-10 17-11 17-11 17-12 17-13 17-14 17-15 17-15 17-16 17-16 17-17

Seeing External Changes .............................................................................................................. Visibility versus Detection of External Changes ...................................................................... Summary of Visibility of Internal and External Changes ....................................................... Oracle Implementation of Scroll-Sensitive Result Sets............................................................

17-17 17-18 17-18 17-19

18

JDBC RowSets
Overview ................................................................................................................................................. RowSet Properties ........................................................................................................................... Events and Event Listeners............................................................................................................ Command Parameters and Command Execution...................................................................... Traversing RowSets ........................................................................................................................ CachedRowSet ....................................................................................................................................... JdbcRowSet ............................................................................................................................................ WebRowSet........................................................................................................................................... FilteredRowSet .................................................................................................................................... JoinRowSet ........................................................................................................................................... 18-1 18-2 18-3 18-4 18-4 18-6 18-9 18-10 18-12 18-13

19

Globalization Support
Providing Globalization Support ...................................................................................................... 19-1 NCHAR, NVARCHAR2, NCLOB and the defaultNChar Property in JDK 1.5 ......................... 19-3 New Methods for National Character Set Type Data in JDK 1.6 ................................................. 19-4

Part V 20

Performance and Scalability

Statement and Resultset Caching

About Statement Caching .................................................................................................................... Basics of Statement Caching .......................................................................................................... Implicit Statement Caching ........................................................................................................... Explicit Statement Caching............................................................................................................ Using Statement Caching .................................................................................................................... Enabling and Disabling Statement Caching................................................................................ Closing a Cached Statement .......................................................................................................... Using Implicit Statement Caching ................................................................................................ Using Explicit Statement Caching ................................................................................................ Reusing Statements Objects................................................................................................................ Using a Pooled Statement .............................................................................................................. Closing a Pooled Statement ........................................................................................................... Resultset Caching .................................................................................................................................. Server-side Cache .......................................................................................................................... Client Query Cache....................................................................................................................... 20-1 20-2 20-2 20-3 20-3 20-4 20-5 20-5 20-7 20-8 20-8 20-9 20-9 20-10 20-10

21

Implicit Connection Caching

The Implicit Connection Cache .......................................................................................................... 21-2 Using the Connection Cache ............................................................................................................... 21-3 Turning Caching On ....................................................................................................................... 21-3

xi

Opening a Connection.................................................................................................................... Setting Connection Cache Name .................................................................................................. Setting Connection Cache Properties ........................................................................................... Closing A Connection..................................................................................................................... Implicit Connection Cache Example ............................................................................................ Connection Attributes .......................................................................................................................... Getting Connections ....................................................................................................................... Setting Connection Attributes ....................................................................................................... Checking Attributes of a Returned Connection ......................................................................... Connection Attribute Example ..................................................................................................... Connection Cache Properties .............................................................................................................. Limit Properties ............................................................................................................................... TIMEOUT Properties.................................................................................................................... Other Properties ............................................................................................................................ Connection Property Example .................................................................................................... Connection Cache Manager API ...................................................................................................... Advanced Topics .................................................................................................................................. Attribute Weights And Connection Matching.......................................................................... Connection Cache Callbacks ....................................................................................................... Use Cases for TimeToLiveTimeout and AbandonedConnectionTimeout ...........................

21-4 21-4 21-5 21-5 21-5 21-6 21-7 21-7 21-8 21-8 21-8 21-9 21-10 21-10 21-11 21-12 21-12 21-12 21-13 21-14

22

Run-Time Connection Load Balancing

Overview ................................................................................................................................................. 22-1 Run-Time Connection Load Balancing ............................................................................................. 22-1 Enabling Run-Time Connection Load Balancing ........................................................................... 22-2

23

Performance Extensions
Update Batching .................................................................................................................................... Overview of Update Batching Models......................................................................................... Oracle Update Batching ................................................................................................................. Oracle Update Batching Characteristics and Limitations .................................................. Setting the Connection Batch Value ...................................................................................... Setting the Statement Batch Value......................................................................................... Checking the Batch Value....................................................................................................... Overriding the Batch Value ................................................................................................... Committing the Changes in Oracle Batching ...................................................................... Update Counts in Oracle Batching........................................................................................ Error Reporting in Oracle Update Batching......................................................................... Standard Update Batching............................................................................................................. Limitations in the Oracle Implementation of Standard Batching..................................... Adding Operations to the Batch ............................................................................................ Processing the Batch .............................................................................................................. Committing the Changes in the Oracle Implementation of Standard Batching........... Clearing the Batch.................................................................................................................. Update Counts in the Oracle Implementation of Standard Batching ............................ Error Handling in the Oracle Implementation of Standard Batching............................ Intermixing Batched Statements and Non-Batched Statements ..................................... 23-1 23-2 23-3 23-3 23-4 23-4 23-5 23-5 23-6 23-6 23-8 23-8 23-8 23-9 23-10 23-10 23-10 23-11 23-12 23-13

xii

Premature Batch Flush ................................................................................................................. Additional Oracle Performance Extensions ................................................................................... Oracle Row-Prefetching Limitations .......................................................................................... Defining Column Types ............................................................................................................... DatabaseMetaData TABLE_REMARKS Reporting..................................................................

23-14 23-15 23-15 23-16 23-19

24

OCI Connection Pooling

OCI Driver Connection Pooling: Background................................................................................. OCI Driver Connection Pooling and Shared Servers Compared ................................................ Defining an OCI Connection Pool ..................................................................................................... Connecting to an OCI Connection Pool ............................................................................................ Sample Code for OCI Connection Pooling ...................................................................................... Statement Handling and Caching ...................................................................................................... JNDI and the OCI Connection Pool ................................................................................................ 24-1 24-2 24-2 24-6 24-7 24-9 24-10

25

Oracle Advanced Queuing

Functionality and Framework ............................................................................................................. AQ Asynchronous Event Notification .............................................................................................. Creating Messages............................................................................................................................... Enqueuing Messages .......................................................................................................................... Dequeuing Messages .......................................................................................................................... Examples: Enqueuing and Dequeuing............................................................................................ 25-1 25-2 25-10 25-12 25-13 25-15

Part VI 26

High Availability

Fast Connection Failover

Overview of Fast Connection Failover .............................................................................................. Using Fast Connection Failover .......................................................................................................... Fast Connection Failover Prerequisites........................................................................................ Configuring ONS For Fast Connection Failover ........................................................................ ONS Configuration File .......................................................................................................... Client-Side ONS Configuration ............................................................................................. Server-Side ONS Configuration Using racgons .................................................................. Remote ONS Subscription ...................................................................................................... Enabling Fast Connection Failover............................................................................................... Querying Fast Connection Failover Status.................................................................................. Understanding Fast Connection Failover ......................................................................................... What The Application Sees............................................................................................................ How It Works .................................................................................................................................. Comparison of Fast Connection Failover and TAF ......................................................................... 26-1 26-2 26-2 26-2 26-2 26-4 26-4 26-5 26-6 26-7 26-7 26-7 26-7 26-8

27

Transparent Application Failover

Overview ................................................................................................................................................. 27-1 Failover Type Events ............................................................................................................................. 27-1 TAF Callbacks ........................................................................................................................................ 27-2

xiii

Java TAF Callback Interface ................................................................................................................ 27-2

Part VII 28

Transaction Management

Distributed Transactions
Overview ................................................................................................................................................. Distributed Transaction Components and Scenarios ................................................................ Distributed Transaction Concepts ................................................................................................ Switching Between Global and Local Transactions ................................................................... Oracle XA Packages ........................................................................................................................ XA Components ..................................................................................................................................... XADatasource Interface and Oracle Implementation................................................................ XAConnection Interface and Oracle Implementation ............................................................... XAResource Interface and Oracle Implementation ................................................................... OracleXAResource Method Functionality and Input Parameters ........................................... Xid Interface and Oracle Implementation ................................................................................. Error Handling and Optimizations .................................................................................................. XAException Classes and Methods............................................................................................ Mapping between Oracle Errors and XA Errors ...................................................................... XA Error Handling........................................................................................................................ Oracle XA Optimizations ............................................................................................................. Implementing a Distributed Transaction ....................................................................................... Summary of Imports for Oracle XA ........................................................................................... Oracle XA Code Sample............................................................................................................... Native-XA in Oracle JDBC Drivers.................................................................................................. OCI Native XA............................................................................................................................... Thin Native XA.............................................................................................................................. 28-1 28-2 28-2 28-4 28-5 28-5 28-6 28-6 28-7 28-8 28-12 28-13 28-13 28-14 28-14 28-15 28-15 28-15 28-15 28-20 28-20 28-21

Part VIII 29

Manageability

Database Management
Database Startup and Shutdown ....................................................................................................... 29-1 Database Change Notification ............................................................................................................ 29-4

30

Diagnosability in JDBC
Logging .................................................................................................................................................... Enabling and Using JDBC Logging .............................................................................................. Configuring the CLASSPATH Environment Variable ....................................................... Enabling Logging..................................................................................................................... Configuring Logging ............................................................................................................... Using Loggers........................................................................................................................... An Example............................................................................................................................... Performance, Scalability, and Security Issues............................................................................. Diagnosability Management............................................................................................................... 30-1 30-2 30-2 30-2 30-3 30-5 30-6 30-7 30-8

xiv

31

JDBC DMS Metrics

Overview ................................................................................................................................................. Determining the Type of Metric to be Generated ........................................................................... Generating the SQLText Metric .......................................................................................................... Accessing DMS Metrics Using JMX .................................................................................................. 31-1 31-2 31-2 31-3

Part IX A

Appendixes

Reference Information
Valid SQL-JDBC Data Type Mappings ............................................................................................... Supported SQL and PL/SQL Data Types ............................................................................................ Embedded SQL92 Syntax ...................................................................................................................... Time and Date Literals ..................................................................................................................... Date Literals................................................................................................................................ Time Literals ............................................................................................................................... Timestamp Literals .................................................................................................................... Scalar Functions................................................................................................................................. LIKE Escape Characters ................................................................................................................... Outer Joins........................................................................................................................................ Function Call Syntax....................................................................................................................... SQL92 to SQL Syntax Example ..................................................................................................... Oracle JDBC Notes and Limitations .................................................................................................. CursorName..................................................................................................................................... SQL92 Outer Join Escapes.............................................................................................................. PL/SQL TABLE, BOOLEAN, and RECORD Types .................................................................. IEEE 754 Floating Point Compliance............................................................................................ Catalog Arguments to DatabaseMetaData Calls........................................................................ SQLWarning Class .......................................................................................................................... Binding Named Parameters .......................................................................................................... A-1 A-3 A-6 A-7 A-7 A-7 A-8 A-9 A-9 A-10 A-10 A-10 A-11 A-11 A-11 A-11 A-12 A-12 A-12 A-12

Coding Tips
JDBC and Multithreading ..................................................................................................................... Performance Optimization .................................................................................................................... Disabling Auto-Commit Mode ....................................................................................................... Standard Fetch Size and Oracle Row Prefetching ........................................................................ Standard and Oracle Update Batching .......................................................................................... Statement Caching ............................................................................................................................ Mapping Between Built-in SQL and Java Types .......................................................................... Transaction Isolation Levels and Access Modes ............................................................................... B-1 B-1 B-1 B-2 B-2 B-3 B-3 B-4

JDBC Error Messages

General Structure of JDBC Error Messages ....................................................................................... General JDBC Messages ........................................................................................................................ JDBC Messages Sorted by ORA Number ...................................................................................... JDBC Messages Sorted in Alphabetic Order ................................................................................. C-1 C-1 C-1 C-5

xv

Native XA Messages ............................................................................................................................... C-9 Native XA Messages Sorted by ORA Number ............................................................................. C-9 Native XA Messages Sorted in Alphabetic Order ........................................................................ C-9 TTC Messages ........................................................................................................................................ C-10 TTC Messages Sorted by ORA Number ...................................................................................... C-10 TTC Messages Sorted in Alphabetic Order ................................................................................. C-11

Troubleshooting
Common Problems .................................................................................................................................. Memory Consumption for CHAR Columns Defined as OUT or IN/OUT Variables ............ Memory Leaks and Running Out of Cursors................................................................................ Boolean Parameters in PL/SQL Stored Procedures..................................................................... Opening More Than 16 OCI Connections for a Process.............................................................. Using statement.cancel ..................................................................................................................... Using JDBC with Firewalls .............................................................................................................. Basic Debugging Procedures ................................................................................................................ Oracle Net Tracing to Trap Network Events ................................................................................ Client-Side Tracing ................................................................................................................... TRACE_LEVEL_CLIENT ................................................................................................. TRACE_DIRECTORY_CLIENT ....................................................................................... TRACE_FILE_CLIENT ...................................................................................................... TRACE_UNIQUE_CLIENT ............................................................................................. Server-Side Tracing ................................................................................................................... TRACE_LEVEL_SERVER ................................................................................................. TRACE_DIRECTORY_SERVER ...................................................................................... TRACE_FILE_SERVER ...................................................................................................... Third Party Debugging Tools ......................................................................................................... D-1 D-1 D-2 D-2 D-2 D-3 D-3 D-4 D-4 D-4 D-4 D-5 D-5 D-5 D-6 D-6 D-6 D-6 D-6

Index

xvi

xvii

List of Tables
11 12 21 31 32 33 41 42 43 44 45 46 47 48 61 62 81 82 83 91 92 93 111 112 121 131 171 181 201 202 203 231 261 281 282 291 292 293 A1 A2 A3 A4 A5 B1 Feature Differences Between JDBC OCI and JDBC Thin Drivers....................................... 1-4 Feature List.................................................................................................................................. 1-5 Import Statements for JDBC Driver ........................................................................................ 2-8 Key Areas of JDBC 3.0 Functionality ...................................................................................... 3-2 BLOB Method Equivalents ....................................................................................................... 3-5 CLOB Method Equivalents....................................................................................................... 3-6 Oracle Data Type Classes ......................................................................................................... 4-5 Key Interfaces and Classes of the oracle.jdbc Package...................................................... 4-19 PL/SQL Types and Corresponding JDBC Types............................................................... 4-29 Arguments of the setPlsqlIndexTable Method................................................................... 4-30 Arguments of the registerIndexTableOutParameter Method .......................................... 4-31 Argument of the getPlsqlIndexTable Method .................................................................... 4-32 Argument of the getOraclePlsqlIndexTable Method ........................................................ 4-33 Arguments of the getPlsqlIndexTable Method .................................................................. 4-33 OCI Instant Client Shared Libraries ........................................................................................ 6-2 Data Shared Library for Instant Client and Instant Client Light (English) ....................... 6-8 Standard Data Source Properties............................................................................................. 8-3 Oracle Extended Data Source Properties................................................................................ 8-4 Supported Database Specifiers ............................................................................................. 8-11 Client/Server Negotiations for Encryption or Integrity ...................................................... 9-7 OCI Driver Client Parameters for Encryption and Integrity ............................................... 9-8 Thin Driver Client Parameters for Encryption and Integrity .............................................. 9-8 Default Mappings Between SQL Types and Java Types................................................... 11-2 getObject and getOracleObject Return Types..................................................................... 11-7 LONG and LONG RAW Data Conversions ....................................................................... 12-3 JPublisher SQL Type Categories, Supported Settings, and Defaults ............................ 13-33 Visibility of Internal and External Changes for Oracle JDBC......................................... 17-19 The JDBC and Cached Row Sets Compared....................................................................... 18-9 Comparing Methods Used in Statement Caching.............................................................. 20-3 Methods Used in Statement Allocation and Implicit Statement Caching ...................... 20-7 Methods Used to Retrieve Explicitly Cached Statements................................................. 20-8 Valid Column Type Specifications ..................................................................................... 23-19 onsctl commands..................................................................................................................... 26-4 Connection Mode Transitions ............................................................................................... 28-4 Oracle-XA Error Mapping ................................................................................................... 28-14 Supported Database Startup Options .................................................................................. 29-2 Supported Database Shutdown Options............................................................................. 29-2 Database Change Notification Registration Options......................................................... 29-6 Valid SQL Data Type-Java Class Mappings ......................................................................... A-1 Support for SQL Data Types ................................................................................................... A-3 Support for ANSI-92 SQL Data Types ................................................................................... A-4 Support for SQL User-Defined Types.................................................................................... A-4 Support for PL/SQL Data Types ............................................................................................ A-5 Mapping of SQL Data Types to Java Classes that Represent SQL Data Types ............... B-3

xviii

Preface
This preface introduces you to the Oracle Database JDBC Developer's Guide and Reference discussing the intended audience, structure, and conventions of this document. A list of related Oracle documents is also provided.

Audience
The Oracle Database JDBC Developer's Guide and Reference is intended for developers of Java Database Connectivity (JDBC)-based applications and applets. This book can be read by anyone with an interest in JDBC programming, but assumes at least some prior knowledge of the following:

Java PL/SQL Oracle databases

Documentation Accessibility
Our goal is to make Oracle products, services, and supporting documentation accessible, with good usability, to the disabled community. To that end, our documentation includes features that make information available to users of assistive technology. This documentation is available in HTML format, and contains markup to facilitate access by the disabled community. Accessibility standards will continue to evolve over time, and Oracle is actively engaged with other market-leading technology vendors to address technical obstacles so that our documentation can be accessible to all of our customers. For more information, visit the Oracle Accessibility Program Web site at http://www.oracle.com/accessibility/ Accessibility of Code Examples in Documentation Screen readers may not always correctly read the code examples in this document. The conventions for writing code require that closing braces should appear on an otherwise empty line; however, some screen readers may not always read a line of text that consists solely of a bracket or brace. Accessibility of Links to External Web Sites in Documentation This documentation may contain links to Web sites of other companies or organizations that Oracle does not own or control. Oracle neither evaluates nor makes any representations regarding the accessibility of these Web sites.

xix

TTY Access to Oracle Support Services Oracle provides dedicated Text Telephone (TTY) access to Oracle Support Services within the United States of America 24 hours a day, 7 days a week. For TTY support, call 800.446.2398. Outside the United States, call +1.407.458.2479.

Related Documents
The following books are also available from the Oracle Java Platform group:

Oracle Database Java Developer's Guide This book introduces the basic concepts of Java and provides general information about server-side configuration and functionality. Information that pertains to the Oracle Java platform as a whole, rather than to a particular product (such as JDBC) is in this book. This book also discusses Java stored procedures, which were formerly discussed in a standalone book.

Oracle Database JPublisher User's Guide This book describes how to use the Oracle JPublisher utility to translate object types and other user-defined types to Java classes. If you are developing JDBC applications that use object types, VARRAY types, nested table types, or object reference types, then JPublisher can generate custom Java classes to map to them.

The following OC4J documents, for Oracle Application Server releases, are also available from the Oracle Java Platform group:

Oracle Application Server Containers for J2EE Users Guide This book provides some overview and general information for OC4J; primer chapters for servlets, JSP pages, and EJBs; and general configuration and deployment instructions.

Oracle Application Server Containers for J2EE Support for JavaServer Pages Developers Guide This book provides information for JSP developers who want to run their pages in OC4J. It includes a general overview of JSP standards and programming considerations, as well as discussion of Oracle value-added features and steps for getting started in the OC4J environment.

Oracle Application Server Containers for J2EE JSP Tag Libraries and Utilities Reference This book provides conceptual information and detailed syntax and usage information for tag libraries, JavaBeans, and other Java utilities provided with OC4J.

Oracle Application Server Containers for J2EE Servlet Developers Guide This book provides information for servlet developers regarding use of servlets and the servlet container in OC4J. It also documents relevant OC4J configuration files.

Oracle Application Server Containers for J2EE Services Guide This book provides information about basic Java services supplied with OC4J, such as JTA, JNDI, and the Oracle Application Server Java Object Cache.

Oracle Application Server Containers for J2EE Enterprise JavaBeans Developers Guide This book provides information about the EJB implementation and EJB container in OC4J.

The following documents are from the Oracle Server Technologies group:
xx

Oracle Database Advanced Application Developer's Guide Oracle Database PL/SQL Packages and Types Reference Oracle Database PL/SQL Language Reference Oracle Database SQL Language Reference Oracle Database Net Services Administrator's Guide Oracle Database Advanced Security Administrator's Guide Oracle Database Reference Oracle Database Error Messages

The following documents from the Oracle Application Server group may also be of some interest:

Oracle Application Server 10g Administrators Guide Oracle Fusion Middleware Administrator's Guide for Oracle HTTP Server Oracle Application Server 10g Performance Guide Oracle Application Server 10g Globalization Guide Oracle Application Server Web Cache Administrators Guide Oracle Application Server 10g Upgrading to 10g (9.0.4)

The following are available from the JDeveloper group:

Oracle JDeveloper online help Oracle JDeveloper documentation on the Oracle Technology Network:
http://otn.oracle.com/products/jdev/content.html

Printed documentation is available for sale in the Oracle Store at: http://oraclestore.oracle.com/ To download free release notes, installation documentation, white papers, or other collateral, visit the Oracle Technology Network (OTN). You must register online before using OTN; registration is free and can be done at http://otn.oracle.com/membership/ If you already have a user name and password for OTN, then you can go directly to the documentation section of the OTN Web site at http://otn.oracle.com/documentation/ The following resources are available from Sun Microsystems:

Web site for Java Platform, Standard Edition (Java SE): http://java.sun.com/javase/technologies/core.jsp

Web site for JDBC, including the latest specifications: http://java.sun.com/javase/technologies/database.jsp

jdbc-interest discussion group for JDBC To subscribe, send an e-mail to listserv@java.sun.com with the following line in the body of the message:
subscribe jdbc-interest yourlastname yourfirstname

xxi

We recommend that you request only the daily digest of the posted e-mails. To do this add the following line to the message body as well:
set jdbc-interest digest

Conventions
This section describes the conventions used in the text and code examples of this documentation set. The following table describes those conventions and provides examples of their use.
Convention Bold Meaning Example

Bold typeface indicates terms that are When you specify this clause, you create an defined in the text or terms that appear in a index-organized table. glossary, or both. Italic typeface indicates book titles or emphasis. Oracle Database Concepts Ensure that the recovery catalog and target database do not reside on the same disk. You can specify this clause only for a NUMBER column. You can back up the database by using the BACKUP command. Query the TABLE_NAME column in the USER_TABLES data dictionary view. Use the DBMS_STATS.GENERATE_STATS procedure. Enter sqlplus to start SQL*Plus. The password is specified in the orapwd file. Back up the datafiles and control files in the /disk1/oracle/dbs directory. The department_id, department_name, and location_id columns are in the hr.departments table. Set the QUERY_REWRITE_ENABLED initialization parameter to true. Connect as oe user. The JRepUtil class implements these methods.

Italics

UPPERCASE monospace (fixed-width) font

Uppercase monospace typeface indicates elements supplied by the system. Such elements include parameters, privileges, data types, RMAN keywords, SQL keywords, SQL*Plus or utility commands, packages and methods, as well as system-supplied column names, database objects and structures, user names, and roles. Lowercase monospace typeface indicates executables, filenames, directory names, and sample user-supplied elements. Such elements include computer and database names, net service names, and connect identifiers, as well as user-supplied database objects and structures, column names, packages and classes, user names and roles, program units, and parameter values. Note: Some programmatic elements use a mixture of UPPERCASE and lowercase. Enter these elements as shown.

lowercase monospace (fixed-width) font

lowercase italic monospace (fixed-width) font

Lowercase italic monospace font represents You can specify the parallel_clause. placeholders or variables. Run old_release.SQL where old_release refers to the release you installed prior to upgrading.

xxii

Whats New
The changes in Oracle Database 11g Release 1 (11.1) can be divided into the following categories:

New Features for Release 1 (11.1) Desupported Features Interface Changes

New Features for Release 1 (11.1)

In this release, Oracle JDBC drivers support the following new features:

JDK 1.5 and 1.6 Support

The Oracle JDBC Thin and OCI drivers have been enhanced to provide support for JDK 1.5 and 1.6. The server-side JDBC drivers provide support for only JDK 1.5.

JDBC 4.0 Support

Oracle JDBC drivers provide support for most of the JDBC 4.0 standard features. Refer to "Support for JDBC 4.0 Standard" on page 3-6 for more information about the JDBC 4.0 standard features

ANYTYPE and ANYDATA

This release of Oracle JDBC drivers provides a Java interface to access SYS.ANYTYPE and SYS.ANYDATA Oracle types. For more information refer "Oracle SYS.ANYTYPE and SYS.ANYDATA Types" Types on page 4-16.

Oracle Advanced Security

Oracle Advanced Security provides confidentiality, integrity, and availability features. This release of Oracle JDBC drivers have been enhanced to support all the features of Oracle Advanced Security. This feature is covered in Chapter 9, "JDBC Client-Side Security Features".

Oracle SecureFiles
Java/JDBC applications get richer SecureFiles LOB data manipulation API and performance enhancements such as versioning, sliding inserts, sliding delete, fragment move, in-place data replacement, compression, encryption, sharing, and client-side read. This feature is covered in "Oracle SecureFiles" on page 14-19.

xxiii

Native Streams AQ Protocol

This release of Oracle JDBC drivers provides a Java interface to Oracle Streams Advanced Queuing (AQ). This feature is covered in Chapter 25, "Oracle Advanced Queuing".

Database Startup and Shutdown

Starting from this release, you can start up and shut down an Oracle Database instance from your JDBC application in the same way as you would from SQL*Plus. This feature is covered in "Database Startup and Shutdown" on page 29-1.

Database Diagnosability
In this release the JDBC drivers have been enhanced by including new diagnosabilty features and improving existing diagnosabilty features. These features enable users to diagnose problems in the applications that use Oracle JDBC drivers and the problems in the drivers themselves. This feature is covered in detail in Chapter 30, "Diagnosability in JDBC".

Database Change Notification

This release Oracle JDBC drivers provide support for the Database Change Notification feature of Oracle Database. Using this functionality of the JDBC drivers, multi-tier systems can take advantage of the Database Change Notification feature to maintain a data cache as updated as possible by receiving invalidation events from the JDBC drivers. This feature is covered in detail in "Database Change Notification" on page 29-4.

New JDBC DMS Metrics with JMX Support

The Dynamic Monitoring Service (DMS) metrics generated in Oracle JDBC 11.1 release are different from 10.2, 10.1, 9.2, and earlier versions of Oracle JDBC as it makes no attempt to retain compatibility with earlier versions. For more information refer "Accessing DMS Metrics Using JMX" on page 31-3.

RowSets in the server

Starting from this release, RowSets are also supported in the server-side drivers, in addition to the Thin and OCI drivers. For more information refer "Overview" on page 18-1.

Desupported Features
From this release onwards, Oracle JDBC drivers will not support JDK versions earlier than 1.5.

Interface Changes
In this release, the oracle.jdbc.OracleConnection interface has been enhanced. For better visibility and clarity, all connection properties are defined as constants in this interface. The oracle.jdbc.driver package, which was deprecated in Oracle Database release 9.0.1, is desupported in this release. Code having references to this package will not compile and run. You can use oracle.jdbc package instead of this package.

xxiv

Part I
Overview
The chapters in this part introduce the concept of Java Database Connectivity (JDBC) and provide an overview of the Oracle implementation of JDBC. This part provides basic information about installation and configuration of the Oracle client with reference to JDBC drivers. This part also covers the basic steps in creating and running any JDBC application. Part I contains the following chapters:

Chapter 1, "Introducing JDBC" Chapter 2, "Getting Started"

1
Introducing JDBC
Java Database Connectivity (JDBC) is a Java standard that provides the interface for connecting from Java to relational databases. The JDBC standard is defined by Sun Microsystems and implemented through the standard java.sql interfaces. This allows individual providers to implement and extend the standard with their own JDBC drivers. JDBC is based on the X/Open SQL Call Level Interface (CLI). JDBC 4.0 complies with the SQL 2003 standard. This chapter provides an overview of the Oracle implementation of JDBC, covering the following topics:

Overview of Oracle JDBC Drivers Environments and Support Feature List

Overview of Oracle JDBC Drivers

In addition to supporting the standard JDBC application programming interfaces (APIs), Oracle drivers have extensions to support Oracle-specific data types and to enhance performance. Oracle provides the following JDBC drivers:

Thin driver It is a pure Java driver used on the client-side, without an Oracle client installation. It can be used with both applets and applications.

Oracle Call Interface (OCI) driver It is used on the client-side with an Oracle client installation. It can be used only with applications.

Server-side Thin driver It is functionally similar to the client-side Thin driver. However, it is used for code that runs on the database server and needs to access another session either on the same server or on a remote server on any tier.

Server-side internal driver It is used for code that runs on the database server and accesses the same session. That is, the code runs and accesses data from a single Oracle session.

Figure 11 illustrates the architecture of Oracle JDBC drivers and Oracle Database.

Introducing JDBC 1-1

Overview of Oracle JDBC Drivers

Figure 11 Architecture of Oracle JDBC Drivers and Oracle Database

Oracle Database JDBC Thin Driver Java Sockets Java Engine Server-Side Thin Driver JDBC Server-Side Internal Driver

JDBC OCI Driver OCI C Library

SOL Engine PL/SQL Engine

KPRB C Library

Oracle Database

This section covers the following topics:

Common Features of Oracle JDBC Drivers Choosing the Appropriate Driver Feature Differences Between JDBC OCI and Thin Drivers

Common Features of Oracle JDBC Drivers

The server-side and client-side Oracle JDBC drivers provide the same basic functionality. The JDBC Thin and OCI drivers support Java Development Kit (JDK) 1.5 and 1.6. The server-side internal drivers support only JDK 1.5. All the JDBC drivers support the following standards and features:

Same syntax and APIs Same Oracle extensions Full support for multithreaded applications

Oracle JDBC drivers implement the standard java.sql interfaces. You can access the Oracle-specific features, in addition to the standard features, by using the oracle.jdbc package. JDBC Thin Driver The JDBC Thin driver is a pure Java, Type IV driver that can be used in applications and applets. It is platform-independent and does not require any additional Oracle software on the client-side. The JDBC Thin driver communicates with the server using SQL*Net to access Oracle Database.

1-2 Oracle Database JDBC Developer's Guide and Reference

Overview of Oracle JDBC Drivers

The JDBC Thin driver allows a direct connection to the database by providing an implementation of SQL*Net on top of Java sockets. The driver supports the TCP/IP protocol and requires a TNS listener on the TCP/IP sockets on the database server.
See Also:

Chapter 5, "Features Specific to JDBC Thin"

JDBC OCI Driver The JDBC OCI driver is a Type II driver used with Java applications. It requires an Oracle client installation and, therefore, is Oracle platform-specific. It supports all installed Oracle Net adapters, including interprocess communication (IPC), named pipes, TCP/IP, and Internetwork Packet Exchange/Sequenced Packet Exchange (IPX/SPX). The JDBC OCI driver, written in a combination of Java and C, converts JDBC invocations to calls to OCI, using native methods to call C-entry points. These calls communicate with the database using SQL*Net. The JDBC OCI driver uses the OCI libraries, C-entry points, Oracle Net, core libraries, and other necessary files on the client computer where it is installed. OCI is an API that enables you to create applications that use the native procedures or function calls of a third-generation language to access Oracle Database and control all phases of the SQL statement processing.
See Also:

Chapter 6, "Features Specific to JDBC OCI Driver"

JDBC Server-Side Thin Driver The JDBC server-side Thin driver offers the same functionality as the JDBC Thin driver that runs on the client-side. However, the JDBC server-side Thin driver runs inside Oracle Database and accesses a remote database or a different session on the same database. This driver is useful in the following scenarios:

Accessing a remote database server from an Oracle Database instance acting as a middle tier Accessing an Oracle Database session from inside another, such as from a Java stored procedure

The use of JDBC Thin driver from a client application or from inside a server does not affect the code.
See Also:

Chapter 5, "Features Specific to JDBC Thin"

JDBC Server-Side Internal Driver The JDBC server-side internal driver supports any Java code that runs inside Oracle Database, such as in a Java stored procedure, and must access the same database. It lets the Java Virtual Machine (JVM) to communicate directly with the SQL engine. This driver supports only JDK 1.5. The JDBC server-side internal driver, the Oracle JVM, the database, and the SQL engine all run within the same address space, and therefore, the issue of network round-trips is irrelevant. The programs access the SQL engine by using function calls.
Note:

The server-side internal driver does not support the cancel and setQueryTimeout methods of the Statement class.

Introducing JDBC 1-3

Environments and Support

The JDBC server-side internal driver is fully consistent with the client-side drivers and supports the same features and extensions.
See Also:

Chapter 7, "Server-Side Internal Driver"

Choosing the Appropriate Driver

Consider the following when choosing a JDBC driver for your application or applet:

In general, unless you need OCI-specific features, such as support for non-TCP/IP networks, use the JDBC Thin driver. If you want maximum portability and performance, then use the JDBC Thin driver. You can connect to Oracle Database from either an application or an applet using the JDBC Thin driver. If you want to use Lightweight Directory Access Protocol (LDAP) over Secure Sockets Layer (SSL), then use the JDBC Thin driver. If you are writing a client application for an Oracle client environment and need OCI-driver-specific features, such as support for non-TCP/IP networks, then use the JDBC OCI driver. If you are writing an applet, then you must use the JDBC Thin driver. For code that runs in the database server and needs to access a remote database or another session within the same database instance, use the JDBC server-side Thin driver. If your code runs inside the database server and needs to access data locally within the session, then use the JDBC server-side internal driver to access that server.

Feature Differences Between JDBC OCI and Thin Drivers

Table 11 lists the features that are specific either to the JDBC OCI or JDBC Thin driver in Oracle Database 11g Release 1 (11.1).
Table 11 Feature Differences Between JDBC OCI and JDBC Thin Drivers JDBC Thin Driver Default support for Native XA

JDBC OCI Driver OCI connection pooling Transparent Application Failover (TAF)

Note:

The OCI optimized fetch and client-side object cache features are internal to the JDBC OCI driver and are not applicable to the JDBC Thin driver. Most JDBC OCI driver features are not available in the JDBC Thin driver because they are inherited from OCI.

Environments and Support

This section provides a brief discussion of the following topics:

Supported JDK and JDBC Versions

1-4 Oracle Database JDBC Developer's Guide and Reference

Feature List

JNI and Java Environments JDBC and IDEs

Supported JDK and JDBC Versions

In Oracle Database 11g Release 1 (11.1), all the JDBC drivers are compatible with JDK 1.5. The JDBC Thin and OCI drivers also support JDK 1.6. All versions of JDK earlier than 1.5 are no longer supported. Support for JDK 1.5 and 1.6 is provided through the ojdbc5.jar and ojdbc6.jar files, respectively.
See Also: "Version Compatibility for Oracle JDBC Drivers" on page 2-1

JNI and Java Environments

The JDBC OCI driver uses the standard Java Native Interface (JNI) to call OCI C libraries. You can use the JDBC OCI driver with JVMs other than that of Sun Microsystems, in particular, with Microsoft and IBM JVMs.

JDBC and IDEs

The Oracle JDeveloper Suite provides developers with a single, integrated set of products to build, debug, and deploy component-based database applications for the Internet. The Oracle JDeveloper environment contains integrated support for JDBC, including the JDBC Thin driver and the native OCI driver. The database component of Oracle JDeveloper uses the JDBC drivers to manage the connection between the application running on the client and the server.

Feature List
Table 12 lists the features and the versions in which they were first supported for each of the three Oracle JDBC drivers: server-side internal driver, JDBC OCI driver, and JDBC Thin driver.
Table 12 Feature JDK 1.0 JDBC 1.0.2 JDK 1.1.1 JDBC 1.22 (No new features; just minor revisions) defineColumnType Row Prefetch Oracle Batching Java Native Interface JDK 1.2 JDBC 2.0 SQL3 Types (BLOB, CLOB, Struct, Array, REF) Native LOB Index-by Tables 10.2.0 9.0.1 8.1.5 Feature List Server-Side Internal JDBC OCI 7.2.2 7.2.2 8.0.6 8.0.6 8.0.6 8.0.6 8.0.6 8.1.6 8.1.6 8.1.5 8.1.6 8.1.6 8.1.6 8.1.5 9.2.0 10.1.0 JDBC Thin 7.2.2 7.2.2 8.0.6 8.0.6 8.0.6 8.0.6 8.0.6

Introducing JDBC 1-5

Feature List

Table 12 (Cont.) Feature List Feature JDBC 2.0 Scrollable ResultSets JDBC 2.0 Updatable ResultSets JDBC 2.0 Standard Batching JDBC 2.0 Connection Pooling JDBC 2.0 XA Server-side Thin driver JDBC 2.0 RowSets Implicit Statement Caching Explicit Statement Caching Temporary LOBs Object Type Inheritance Multilevel Collections oracle.jdbc Interfaces Native XA OCI Connection Pooling TAF NLS Support JDK 1.3 JDK 1.4 JDBC 3.0 Savepoints New Statement Caching API ConnectionCacheImpl connection cache Implicit Connection Cache Fast Connection Failover Connection Wrapping DMS Service Names in URLs JDBC 3.0 Connection Pooling Properties JDBC 3.0 Updatable BLOB, CLOB, REF JDBC 3.0 Multiple Open ResultSets JDBC 3.0 Parameter Metadata JDBC 3.0 Set/Get Stored Procedures Parameters by Name JDBC 3.0 Statement Pooling Set Statement Parameters by Name End-to-End Tracing Web RowSet NA 10.1.0 10.2.0 10.1.0 10.1.0 10.1.0 10.1.0 NA NA 9.0.1 9.2.0 10.1.0 9.2.0 9.2.0 NA NA 8.1.7 8.1.7 9.0.1 9.0.1 9.0.1 9.0.1 Server-Side Internal 8.1.6 8.1.6 8.1.6 NA 8.1.6 8.1.6 JDBC OCI 8.1.6 8.1.6 8.1.6 8.1.6 8.1.6 NA 9.0.1 8.1.7 8.1.7 9.0.1 9.0.1 9.0.1 9.0.1 9.0.1 9.0.1 9.0.1 9.0.1 9.2.0 9.2.0 9.2.0 9.2.0 8.1.7 10.1.0 10.1.0.3 9.2.0 9.2.0 9.2.0 10.1.0 10.1.0 10.2.0 10.1.0 10.1.0 10.1.0 10.1.0 10.1.0 10.1.0 JDBC Thin 8.1.6 8.1.6 8.1.6 8.1.6 8.1.6 NA 9.0.1 8.1.7 8.1.7 9.0.1 9.0.1 9.0.1 9.0.1 10.1.0 NA NA 9.0.1 9.2.0 9.2.0 9.2.0 9.2.0 8.1.7 10.1.0 10.1.0.3 9.2.0 9.2.0 10.2.0 10.1.0 10.1.0 10.2.0 10.1.0 10.1.0 10.1.0 10.1.0 10.1.0 10.1.0

1-6 Oracle Database JDBC Developer's Guide and Reference

Feature List

Table 12 (Cont.) Feature List Feature Proxy Authentication JDBC 3.0 Auto Generated Keys JDBC 3.0 Holdable Cursors JDBC 3.0 Local/Global Transaction Switching Run-time Connection Load Balancing Extended setXXX and getXXX for LOBs XA Connection Cache DML Returning JSR 114 RowSets SSL Encryption SSL Authentication Radius Authentication JDK 1.5 JDK 1.6 JDBC 4.0 Database startup and shutdown Java interface to Streams AQ 11.1 NA 10.2 NA 10.2.0 9.2.0 NA Server-Side Internal JDBC OCI 10.2.0 10.2.0 10.2.0 9.2.0 10.2.0 10.2.0 10.2.0 10.2.0 10.2.0 9.2.0 9.2.0 10.2.0 10.2 11.1 11.1 11.1 10.2 11.1 11.1 11.1 11.1 JDBC Thin 10.1.0 10.2.0 10.2.0 9.2.0 10.2.0 10.2.0 10.2.0 10.2.0 10.2.0 10.2.0

Note:

In the table, NA means that the feature is not applicable for the corresponding Oracle JDBC driver. The ConnectionCacheImpl connection cache feature is deprecated in Oracle Database 10g. Implicit Connection Cache replaces this in Oracle Database 10g.

Introducing JDBC 1-7

Feature List

1-8 Oracle Database JDBC Developer's Guide and Reference

2
Getting Started
This chapter discusses the compatibility of Oracle Java Database Connectivity (JDBC) driver versions, database versions, and Java Development Kit (JDK) versions. It also describes the basics of testing a client installation and configuration and running a simple application. This chapter contains the following sections:

Version Compatibility for Oracle JDBC Drivers Verification of a JDBC Client Installation Basic Steps in JDBC Sample: Connecting, Querying, and Processing the Results Stored Procedure Calls in JDBC Programs Processing SQL Exceptions

Version Compatibility for Oracle JDBC Drivers

This section discusses the general JDBC version compatibility issues. Backward Compatibility The JDBC drivers are certified to work with the currently supported versions of Oracle Database. For example, the JDBC Thin drivers in Oracle Database 11g Release 1 (11.1) are certified to work with the 10.2.x, 10.1.x, 9.2.x, and 9.0.1.x Oracle Database releases. However, they are not certified to work with older, unsupported database releases, such as 8.0.x and 7.x. Forward Compatibility Existing and supported JDBC drivers are certified to work with Oracle Database 11g Release 1 (11.1).
Note:

In Oracle Database 11g Release 1 (11.1), Oracle JDBC drivers no longer support JDK 1.4.x or earlier versions. You can find a complete, up-to-date list of supported databases at http://www.oracle.com/technology/tech/java/sqlj _jdbc/htdocs/jdbc_faq.htm.

Getting Started 2-1

Verification of a JDBC Client Installation

Verification of a JDBC Client Installation

To verify a JDBC client installation, you must do all of the following:

Check the Installed Directories and Files Check the Environment Variables Ensure that the Java Code Can Be Compiled and Run Determine the Version of the JDBC Driver Test JDBC and the Database Connection

Installation of an Oracle JDBC driver is platform-specific. Follow the installation instructions for the driver you want to install in your platform-specific documentation. This section describes the steps for verifying an Oracle client installation of the JDBC drivers, assuming that you have already installed the driver of your choice. If you have installed the JDBC Thin driver, then no further installation on the client computer is necessary.
Note:

The JDBC Thin driver requires a TCP/IP listener to be running on the computer where the database is installed.

If you have installed the JDBC Oracle Call Interface (OCI) driver, then you must also install the Oracle client software. This includes Oracle Net and the OCI libraries.

Check the Installed Directories and Files

Installing the Oracle Java products creates, among other things, the following directories:

ORACLE_HOME/jdbc ORACLE_HOME /jlib

Check whether or not the following directories and files have been created and populated in the ORACLE_HOME/jdbc directory:

demo This directory contains a compressed file, demo.zip or demo.tar. When you uncompress this compressed file, the samples directory and the Samples-Readme.txt file are created. The samples directory contains sample programs, including examples of how to use SQL92 and Oracle SQL syntax, PL/SQL blocks, streams, user-defined types, additional Oracle type extensions, and Oracle performance extensions.

doc This directory contains the javadoc.zip file, which is the Oracle JDBC application programming interface (API) documentation.

lib The lib directory contains the following required Java classes: orai18n.jar and orai18n-mapping.jar Contain classes for globalization and multibyte character sets support ojdbc5.jar, ojdbc5_g.jar, ojdbc6.jar, and ojdbc6_g.jar

2-2 Oracle Database JDBC Developer's Guide and Reference

Verification of a JDBC Client Installation

Contain the JDBC driver classes for use with JDK 1.5 and JDK 1.6
Note:

In Oracle Database 11g Release 1 (11.1), support for a version of JDK earlier than version 1.5 has been removed. Also, the ojdbc14.jar and classes12.jar files are no longer shipped. Instead, you can use the ojdbc5.jar and ojdbc6.jar files, which are shipped with Oracle Database 11g. If you are using JSE 6 and later, then there is no need to explicitly load the JDBC driver. This means that the Java run-time loads the driver when needed and you need not include Class.forName("oracle.jdbc.OracleDriver") or new oracle.jdbc.OracleDriver() in your code. But if you are using J2SE 5, then you need to load the JDBC driver explicitly.

Readme.txt This file contains late-breaking and release-specific information about the drivers, which may not have been included in other documentation on the product.

Check whether or not the following directories have been created and populated in the ORACLE_HOME /jlib directory:

jta.jar and jndi.jar These files contain classes for the Java Transaction API (JTA) and the Java Naming and Directory Interface (JNDI). These are required only if you are using JTA features for distributed transaction management or JNDI features for naming services.

Note:

These files can also be obtained from the Sun Microsystems Web site. However, it is recommended that you use the versions supplied by Oracle, which have been tested with the Oracle drivers.

Check the Environment Variables

This section describes the environment variables that must be set for the JDBC OCI driver and the JDBC Thin driver, focusing on the Sun Solaris, Linux, and Microsoft Windows platforms. You must set the CLASSPATH environment variable for your installed JDBC OCI or Thin driver. Include the following in the CLASSPATH environment variable:
ORACLE_HOME/jdbc/lib/ojdbc5.jar ORACLE_HOME/jlib/orai18n.jar

Note:

If you use the JTA features and the JNDI features, then you must specify jta.jar and jndi.jar in your CLASSPATH environment variable.

Getting Started 2-3

Verification of a JDBC Client Installation

JDBC OCI Driver If you are installing the JDBC OCI driver, then you must also set the following value for the library path environment variable:

On Sun Solaris or Linux, set the LD_LIBRARY_PATH environment variable as follows:

ORACLE_HOME/lib

This directory contains the libocijdbc11.so shared object library.

Note:

If you are running a 32-bit Java Virtual Machine (JVM) against a 64-bit client or database, then you must also add ORACLE_HOME/lib32 to the LD_LIBRARY_PATH environment variable.

On Microsoft Windows, set the PATH environment variable as follows:

ORACLE_HOME\bin

This directory contains the ocijdbc11.dll dynamic link library. All of the JDBC OCI demonstration programs can be run in the Instant Client mode by including the JDBC OCI Instant Client data shared library on the library path environment variable.
See Also:

Chapter 6, "Features Specific to JDBC OCI Driver"

JDBC Thin Driver If you are installing the JDBC Thin driver, then you do not have to set any other environment variables. However, to use the JDBC server-side Thin driver, you need to set permission. Setting Permission for the Server-Side Thin Driver The JDBC server-side Thin driver opens a socket for its connection to the database. Because Oracle Database enforces the Java security model, a check is performed for a SocketPermission object. To use the JDBC server-side Thin driver, the connecting user must be granted the appropriate permission. The following is an example of how the permission can be granted for the user SCOTT:
CREATE ROLE jdbcthin; CALL dbms_java.grant_permission('JDBCTHIN', 'java.net.SocketPermission', '*', 'connect'); GRANT jdbcthin TO SCOTT;

Note that JDBCTHIN in the grant_permission call must be in uppercase. The asterisk (*) is a pattern. You can restrict the user by granting permission to connect to only specific computers or ports.
See Also:

Oracle Database Java Developer's Guide

Ensure that the Java Code Can Be Compiled and Run

To further ensure that Java is set up properly on your client system, go to the samples directory under the ORACLE_HOME/jdbc/demo directory. Now, type the following

2-4 Oracle Database JDBC Developer's Guide and Reference

Verification of a JDBC Client Installation

commands on the command line, one after the other, to see if the Java compiler and the Java interpreter run without error. :
javac java

Each of the preceding commands should display a list of options and parameters and then exit. Ideally, verify that you can compile and run a simple test program, such as jdbc/demo/samples/generic/SelectExample.

Determine the Version of the JDBC Driver

You can determine the version of the JDBC driver that you installed, by calling the getDriverVersion method of the OracleDatabaseMetaData class. The following sample code shows how to determine the driver version:
import java.sql.*; import oracle.jdbc.*; import oracle.jdbc.pool.OracleDataSource; class JDBCVersion { public static void main (String args[]) throws SQLException { OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:oracle:thin:scott/tiger@host:port:service"); Connection conn = ods.getConnection(); // Create Oracle DatabaseMetaData object DatabaseMetaData meta = conn.getMetaData(); // gets driver info: System.out.println("JDBC driver version is " + meta.getDriverVersion()); } }

You can also determine the version of the JDBC driver by executing the following commands:

java -jar ojdbc5.jar java -jar ojdbc6.jar

Test JDBC and the Database Connection

The samples directory contains sample programs for a particular Oracle JDBC driver. One of the programs, JdbcCheckup.java, is designed to test JDBC and the database connection. The program queries for the user name, password, and the name of the database to which you want to connect. The program connects to the database, queries for the string "Hello World", and prints it to the screen. Go to the samples directory, and compile and run the JdbcCheckup.java program. If the results of the query print without error, then your Java and JDBC installations are correct. Although JdbcCheckup.java is a simple program, it demonstrates several important functions by performing the following:

Imports the necessary Java classes, including JDBC classes

Getting Started 2-5

Verification of a JDBC Client Installation

Creates a DataSource instance Connects to the database Runs a simple query Prints the query results to your screen

The JdbcCheckup.java program, which uses the JDBC OCI driver, is as follows:
/* * This sample can be used to check the JDBC installation. * Just run it and provide the connect information. It will select * "Hello World" from the database. */ // You import import import need to import the java.sql and JDBC packages to use JDBC java.sql.*; oracle.jdbc.*; oracle.jdbc.pool.OracleDataSource;

// We import java.io to be able to read from the command line import java.io.*; class JdbcCheckup { public static void main(String args[]) throws SQLException, IOException { // Prompt the user for connect information System.out.println("Please enter information to test connection to the database"); String user; String password; String database; user = readEntry("user: "); int slash_index = user.indexOf('/'); if (slash_index != -1) { password = user.substring(slash_index + 1); user = user.substring(0, slash_index); } else password = readEntry("password: "); database = readEntry("database(a TNSNAME entry): "); System.out.print("Connecting to the database..."); System.out.flush(); System.out.println("Connecting..."); // Open an OracleDataSource and get a connection OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:oracle:oci:@" + database); ods.setUser(user); ods.setPassword(password); Connection conn = ods.getConnection(); System.out.println("connected."); // Create a statement Statement stmt = conn.createStatement(); // Do the SQL "Hello World" thing

2-6 Oracle Database JDBC Developer's Guide and Reference

Basic Steps in JDBC

ResultSet rset = stmt.executeQuery("select 'Hello World' from dual"); while (rset.next()) System.out.println(rset.getString(1)); // close the result set, the statement and connect rset.close(); stmt.close(); conn.close(); System.out.println("Your JDBC installation is correct."); } // Utility function to read a line from standard input static String readEntry(String prompt) { try { StringBuffer buffer = new StringBuffer(); System.out.print(prompt); System.out.flush(); int c = System.in.read(); while (c != '\n' && c != -1) { buffer.append((char)c); c = System.in.read(); } return buffer.toString().trim(); } catch(IOException e) { return ""; } } }

Basic Steps in JDBC

After verifying the JDBC client installation, you can start creating your JDBC applications. When using Oracle JDBC drivers, you must include certain driver-specific information in your programs. This section describes, in the form of a tutorial, where and how to add the information. The tutorial guides you through the steps to create code that connects to and queries a database from the client. You must write code to perform the following tasks:
1. 2. 3. 4. 5. 6. 7. 8. 9.

Importing Packages Opening a Connection to a Database Creating a Statement Object Running a Query and Retrieving a ResultSet Object Processing the ResultSet Object Closing the ResultSet and Statement Objects Making Changes to the Database Committing Changes Closing the Connection

Getting Started 2-7

Basic Steps in JDBC

Note:

You must supply Oracle driver-specific information for the first three tasks, which allow your program to use the JDBC application programming interface (API) to access a database. For the other tasks, you can use standard JDBC Java code, as you would for any Java application.

Importing Packages
Regardless of which Oracle JDBC driver you use, include the import statements shown in Table 21 at the beginning of your program.
Table 21 Import Statements for JDBC Driver Provides Standard JDBC packages. The BigDecimal and BigInteger classes. You can omit this package if you are not going to use these classes in your application. Oracle extensions to JDBC. This is optional. OracleDataSource. Oracle type extensions. This is optional.

Import statement import java.sql.*; import java.math.*;

import oracle.jdbc.*; import oracle.jdbc.pool.*; import oracle.sql.*;

The Oracle packages listed as optional provide access to the extended functionality provided by Oracle JDBC drivers, but are not required for the example presented in this section.
Note:

It is better to import only the classes your application needs, rather than using the wildcard asterisk (*). This guide uses the asterisk (*) for simplicity, but this is not the recommended way of importing classes and interfaces.

Opening a Connection to a Database

First, you must create an OracleDataSource instance. Then, open a connection to the database using the OracleDataSource.getConnection method. The properties of the retrieved connection are derived from the OracleDataSource instance. If you set the URL connection property, then all other properties, including TNSEntryName, DatabaseName, ServiceName, ServerName, PortNumber, Network Protocol, and driver type are ignored. Specifying a Database URL, User Name, and Password The following code sets the URL, user name, and password for a data source:
OracleDataSource ods = new OracleDataSource(); ods.setURL(url); ods.setUser(user); ods.setPassword(password);

The following example connects user scott with password tiger to a database with service orcl through port 1521 of the host myhost, using the JDBC Thin driver:
OracleDataSource ods = new OracleDataSource(); String url = "jdbc:oracle:thin:@//myhost:1521/orcl",

2-8 Oracle Database JDBC Developer's Guide and Reference

Basic Steps in JDBC

ods.setURL(url); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection();

Note:

The user name and password specified in the arguments override any user name and password specified in the URL.

Specifying a Database URL that Includes User Name and Password The following example connects user scott with password tiger to a database host whose Transparent Network Substrate (TNS) entry is myTNSEntry, using the JDBC Oracle Call Interface (OCI) driver. In this case, the URL includes the user name and password and is the only input parameter.
String url = "jdbc:oracle:oci:scott/tiger@myTNSEntry"); ods.setURL(url); Connection conn = ods.getConnection();

If you want to connect using the Thin driver, then you must specify the port number. For example, if you want to connect to the database on the host myhost that has a TCP/IP listener on port 1521 and the service identifier is orcl, then provide the following code:
String URL = "jdbc:oracle:thin:scott/tiger@//myhost:1521/orcl"); ods.setURL(URL); Connection conn = ods.getConnection();

See Also:

Chapter 8, "Data Sources and URLs"

Creating a Statement Object

Once you connect to the database and, in the process, create a Connection object, the next step is to create a Statement object. The createStatement method of the JDBC Connection object returns an object of the JDBC Statement type. To continue the example from the previous section, where the Connection object conn was created, here is an example of how to create the Statement object:
Statement stmt = conn.createStatement();

Running a Query and Retrieving a ResultSet Object

To query the database, use the executeQuery method of the Statement object. This method takes a SQL statement as input and returns a JDBC ResultSet object.

Getting Started 2-9

Basic Steps in JDBC

Note:

The method used to execute a Statement object depends on the type of SQL statement being executed. If the Statement object represents a SQL query returning a ResultSet object, the executeQuery method should be used. If the SQL is known to be a DDL statement or a DML statement returning an update count, the executeUpdate method should be used. If the type of the SQL statement is not known, the execute method should be used. In case of a standard JDBC driver, if the SQL string being executed does not return a ResultSet object, then the executeQuery method throws a SQLException exception. In case of an Oracle JDBC driver, the executeQuery method does not throw a SQLException exception even if the SQL string being executed does not return a ResultSet object.

To continue the example, once you create the Statement object stmt, the next step is to run a query that returns a ResultSet object with the contents of the ename column of a table of employees named EMP:
ResultSet rset = stmt.executeQuery ("SELECT ename FROM emp");

Processing the ResultSet Object

Once you run your query, use the next() method of the ResultSet object to iterate through the results. This method steps through the result set row by row, detecting the end of the result set when it is reached. To pull data out of the result set as you iterate through it, use the appropriate getXXX methods of the ResultSet object, where XXX corresponds to a Java data type. For example, the following code will iterate through the ResultSet object, rset, from the previous section and will retrieve and print each employee name:
while (rset.next()) System.out.println (rset.getString(1));

The next() method returns false when it reaches the end of the result set. The employee names are materialized as Java String values.

Closing the ResultSet and Statement Objects

You must explicitly close the ResultSet and Statement objects after you finish using them. This applies to all ResultSet and Statement objects you create when using Oracle JDBC drivers. The drivers do not have finalizer methods. The cleanup routines are performed by the close method of the ResultSet and Statement classes. If you do not explicitly close the ResultSet and Statement objects, serious memory leaks could occur. You could also run out of cursors in the database. Closing both the result set and the statement releases the corresponding cursor in the database. If you close only the result set, then the cursor is not released. For example, if your ResultSet object is rset and your Statement object is stmt, then close the result set and statement with the following lines of code:
rset.close();

2-10 Oracle Database JDBC Developer's Guide and Reference

Basic Steps in JDBC

stmt.close();

When you close a Statement object that a given Connection object creates, the connection itself remains open.
Note:

Typically, you should put close statements in a finally

clause.

Making Changes to the Database

To write changes to the database, such as for INSERT or UPDATE operations, you create a PreparedStatement object. This enables you to run a statement with varying sets of input parameters. The prepareStatement method of the JDBC Connection object lets you define a statement that takes variable bind parameters and returns a JDBC PreparedStatement object with your statement definition. Use the setXXX methods on the PreparedStatement object to bind data to the prepared statement to be sent to the database.
See Also:

"The setObject and setOracleObject Methods" on page 11-11 and "Other setXXX Methods" on page 11-11

The following example shows how to use a prepared statement to run INSERT operations that add two rows to the EMP table.
// Prepare to insert new names in the EMP table PreparedStatement pstmt = null; try{ pstmt = conn.prepareStatement ("insert into EMP (EMPNO, ENAME) values (?, ?)"); // Add LESLIE as employee number 1500 pstmt.setInt (1, 1500); // The first ? is for EMPNO pstmt.setString (2, "LESLIE"); // The second ? is for ENAME // Do the insertion pstmt.execute (); // Add MARSHA as employee number 507 pstmt.setInt (1, 507); // The first ? is for EMPNO pstmt.setString (2, "MARSHA"); // The second ? is for ENAME // Do the insertion pstmt.execute (); } finally{ if(pstmt!=null) // Close the statement pstmt.close(); }

Committing Changes
By default, data manipulation language (DML) operations are committed automatically as soon as they are run. This is known as the auto-commit mode. However, you can disable auto-commit mode with the following method call on the Connection object:
Getting Started 2-11

Sample: Connecting, Querying, and Processing the Results

conn.setAutoCommit(false);

See Also:

"Disabling Auto-Commit Mode" on page B-1.

If you disable the auto-commit mode, then you must manually commit or roll back changes with the appropriate method call on the Connection object:
conn.commit();

or:
conn.rollback();

A COMMIT or ROLLBACK operation affects all DML statements run since the last COMMIT or ROLLBACK.
Note:

If the auto-commit mode is disabled and you close the connection without explicitly committing or rolling back your last changes, then an implicit COMMIT operation is run. Any data definition language (DDL) operation always causes an implicit COMMIT. If the auto-commit mode is disabled, then this implicit COMMIT will commit any pending DML operations that had not yet been explicitly committed or rolled back.

Closing the Connection

You must close the connection to the database after you have performed all the required operations and no longer require the connection. You can close the connection by using the close method of the Connection object, as follows:
conn.close();

Note:

Typically, you should put close statements in a finally

clause.

Sample: Connecting, Querying, and Processing the Results

The steps in the preceding sections are illustrated in the following example, which uses the Oracle JDBC Thin driver to create a data source, connects to the database, creates a Statement object, runs a query, and processes the result set. Note that the code for creating the Statement object, running the query, returning and processing the ResultSet object, and closing the statement and connection uses the standard JDBC API.
import import import import java.sql.Connection; java.sql.Resultset; java.sql.Statement; oracle.jdbc.pool.OracleDataSource;

class JdbcTest { public static void main (String args []) throws SQLException

2-12 Oracle Database JDBC Developer's Guide and Reference

Stored Procedure Calls in JDBC Programs

{ OracleDataSource ods = null; Connection conn = null; Statement stmt = null; ResultSet rset = null; // Create DataSource and connect to the local database ods = new OracleDataSource(); ods.setURL("jdbc:oracle:thin:@//myhost:1521/orcl"); ods.setUser("scott"); ods.setPassword("tiger"); conn = ods.getConnection(); try { // Query the employee names stmt = conn.createStatement (); rset = stmt.executeQuery ("SELECT ename FROM emp"); // Print the name out while (rset.next ()) System.out.println (rset.getString (1)); } //Close the result set, statement, and the connection finally{ if(rset!=null) rset.close(); if(stmt!=null) stmt.close(); if(conn!=null) conn.close(); } } }

If you want to adapt the code for the OCI driver, then replace the call to the OracleDataSource.setURL method with the following:
ods.setURL("jdbc:oracle:oci:@MyHostString");

where, MyHostString is an entry in the TNSNAMES.ORA file.

Stored Procedure Calls in JDBC Programs

This section describes how Oracle JDBC drivers support the following kinds of stored procedures:

PL/SQL Stored Procedures Java Stored Procedures

PL/SQL Stored Procedures

Oracle JDBC drivers support the processing of PL/SQL stored procedures and anonymous blocks. They support PL/SQL block syntax and most of SQL92 escape syntax. The following PL/SQL calls would work with any Oracle JDBC driver:
// SQL92 syntax CallableStatement cs1 = conn.prepareCall ( "{call proc (?,?)}" ) ; // stored proc

Getting Started 2-13

Processing SQL Exceptions

CallableStatement cs2 = conn.prepareCall ( "{? = call func (?,?)}" ) ; // stored func // PL/SQL block syntax CallableStatement cs3 = conn.prepareCall ( "begin proc (?,?); end;" ) ; // stored proc CallableStatement cs4 = conn.prepareCall ( "begin ? := func(?,?); end;" ) ; // stored func

As an example of using the Oracle syntax, here is a PL/SQL code snippet that creates a stored function. The PL/SQL function gets a character sequence and concatenates a suffix to it:
create or replace function foo (val1 char) return char as begin return val1 || 'suffix'; end;

The function invocation in your JDBC program should look like the following:
OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:oracle:oci:@<hoststring>"); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection(); CallableStatement cs = conn.prepareCall ("begin ? := foo(?); end;"); cs.registerOutParameter(1,Types.CHAR); cs.setString(2, "aa"); cs.executeUpdate(); String result = cs.getString(1);

Java Stored Procedures

You can use JDBC to call Java stored procedures through the SQL and PL/SQL engines. The syntax for calling Java stored procedures is the same as the syntax for calling PL/SQL stored procedures, presuming they have been properly published. That is, you have written call specifications to publish them to the Oracle data dictionary. Applications can call Java stored procedures using the Native Java Interface for direct invocation of static Java methods.

Processing SQL Exceptions

To handle error conditions, Oracle JDBC drivers throw SQL exceptions, producing instances of the java.sql.SQLException class or its subclass. Errors can originate either in the JDBC driver or in the database itself. Resulting messages describe the error and identify the method that threw the error. Additional run-time information can also be appended. JDBC 3.0 defines only a single exception, SQLException. However, there are large categories of errors and it is useful to distinguish them. Therefore, in JDBC 4.0, a set of subclasses of the SQLException exception is introduced to identify the different categories of errors. To know more about this feature, see Support for JDBC 4.0 Standard on page 3-6. Basic exception handling can include retrieving the error message, retrieving the error code, retrieving the SQL state, and printing the stack trace. The SQLException class includes functionality to retrieve all of this information, when available.

2-14 Oracle Database JDBC Developer's Guide and Reference

Processing SQL Exceptions

See Also:

Appendix C, "JDBC Error Messages" Oracle Database Error Messages

Retrieving Error Information You can retrieve basic error information with the following methods of the SQLException class:

getMessage getErrorCode getSQLState

The following example prints output from a getMessage method call:

catch(SQLException e) { System.out.println("exception: " + e.getMessage()); }

This would print the output, such as the following, for an error originating in the JDBC driver:
exception: Invalid column type

Note:

Error message text is available in alternative languages and character sets supported by Oracle.

Printing the Stack Trace The SQLException class provides the printStackTrace() method for printing a stack trace. This method prints the stack trace of the throwable object to the standard error stream. You can also specify a java.io.PrintStream object or java.io.PrintWriter object for output. The following code fragment illustrates how you can catch SQL exceptions and print the stack trace.
try { <some code> } catch(SQLException e) { e.printStackTrace (); }

To illustrate how the JDBC drivers handle errors, assume the following code uses an incorrect column index:
// Iterate through the result and print the employee names // of the code try { while (rset.next ()) System.out.println (rset.getString (5)); // incorrect column index } catch(SQLException e) { e.printStackTrace (); }

Assuming the column index is incorrect, running the program would produce the following error text:
java.sql.SQLException: Invalid column index at oracle.jdbc.driver.DatabaseError.throwSqlException(DatabaseError.java:112)

Getting Started 2-15

Processing SQL Exceptions

at at at at

oracle.jdbc.driver.DatabaseError.throwSqlException(DatabaseError.java:146) oracle.jdbc.driver.DatabaseError.throwSqlException(DatabaseError.java:208) oracle.jdbc.driver.OracleResultSetImpl.getDate(OracleResultSetImpl.java:1556) Employee.main(Employee.java:41)

2-16 Oracle Database JDBC Developer's Guide and Reference

Part II
Oracle JDBC
This part includes chapters that discuss the different Java Database Connectivity (JDBC) versions that Oracle Database 10g supports. It also includes chapters that cover features specific to JDBC Thin driver, JDBC Oracle Call Interface (OCI) driver, and the server-side internal driver. Part II contains the following chapters:

Chapter 3, "JDBC Standards Support" Chapter 4, "Oracle Extensions" Chapter 5, "Features Specific to JDBC Thin" Chapter 6, "Features Specific to JDBC OCI Driver" Chapter 7, "Server-Side Internal Driver"

3
JDBC Standards Support
The Oracle Java Database Connectivity (JDBC) drivers support different versions of the JDBC standard features. In Oracle Database 11g Release 1 (11.1), Oracle JDBC drivers have been enhanced to provide support for the JDBC 4.0 standards. These features are provided through the oracle.jdbc and oracle.sql packages. These packages support Java Development Kit (JDK) releases 1.5 and 1.6. This chapter discusses the JDBC standards support in Oracle JDBC drivers. It contains the following sections:

Support for JDBC 2.0 Standard Support for JDBC 3.0 Standard Support for JDBC 4.0 Standard

Support for JDBC 2.0 Standard

Standard JDBC 2.0 features are supported by JDK 1.2 and later versions. There are three areas to consider:

Support for data types, such as objects, arrays, and large objects (LOBs). This is handled through the standard java.sql package. Support for standard features, such as result set enhancements and update batching. This is handled through standard objects, such as Connection, ResultSet, and PreparedStatement, under JDK 1.2.x and later. Support for extended features, such as features of the JDBC 2.0 optional package, also known as the standard extension application programming interface (API), including data sources, connection pooling, and distributed transactions.

This section covers the following topics:

Data Type Support Standard Feature Support Extended Feature Support Standard versus Oracle Performance Enhancement APIs
Note:

Versions of JDK earlier than 1.5 are no longer supported. The package oracle.jdbc2 has been removed.

JDBC Standards Support

3-1

Support for JDBC 3.0 Standard

Data Type Support

Oracle JDBC fully supports JDK 1.5 and JDK 1.6, which includes standard JDBC 2.0 functionality through implementation of interfaces in the standard java.sql package. These interfaces are implemented as appropriate by classes in the oracle.sql and oracle.jdbc packages.

Standard Feature Support

In a JDK 1.5 environment, using the JDBC classes in ojdbc5.jar, JDBC 2.0 features, such as scrollable result sets, updatable result sets, and update batching, are supported through methods specified by standard JDBC 2.0 interfaces.

Extended Feature Support

Features of the JDBC 2.0 optional package, including data sources, connection pooling, and distributed transactions, are supported in a JDK 1.2.x or later environment. The standard javax.sql package and classes that implement its interfaces are included in the Java Archive (JAR) files packaged with Oracle Database.

Standard versus Oracle Performance Enhancement APIs

The following performance enhancements are available under JDBC 2.0, which had previously been available only as Oracle extensions:

Update batching Fetch size or row prefetching

In each case, you have the option of using the standard model or the Oracle model. Oracle recommends that you use the JDBC standard model whenever possible. Do not, however, try to mix usage of the standard model and Oracle model within a single application for either of these features.
See Also:

"Update Batching" on page 23-1 "Fetch Size" on page 17-15

Support for JDBC 3.0 Standard

Standard JDBC 3.0 features are supported by JDK 1.4 and earlier versions. Table 31 lists the JDBC 3.0 features supported by Oracle Database 11g Release 1 (11.1) and gives references to a detailed discussion of each feature.
Table 31 Feature Transaction savepoints Statement caching Key Areas of JDBC 3.0 Functionality Comments and References See "Transaction Savepoints" on page 3-3 for information. Reuse of prepared statements by connection pools. See Chapter 20, "Statement and Resultset Caching".

Switching between local and See "Switching Between Global and Local Transactions" on page 28-4. global transactions LOB modification Named SQL parameters See "JDBC 3.0 LOB Interface Methods" on page 3-5. See "Interface oracle.jdbc.OracleCallableStatement" on page 4-22 and "Interface oracle.jdbc.OraclePreparedStatement" on page 4-22.

3-2 Oracle Database JDBC Developer's Guide and Reference

Support for JDBC 3.0 Standard

Table 31 (Cont.) Key Areas of JDBC 3.0 Functionality Feature RowSets Retrieving auto-generated keys Result set holdability Comments and References See Chapter 18, "JDBC RowSets" See "Retrieval of Auto-Generated Keys" on page 3-4 See "Result Set Holdability" on page 3-6

The following JDBC 3.0 features supported by Oracle JDBC drivers are covered in this section:

Transaction Savepoints Retrieval of Auto-Generated Keys JDBC 3.0 LOB Interface Methods Result Set Holdability

Transaction Savepoints
The JDBC 3.0 specification supports savepoints, which offer finer demarcation within transactions. Applications can set a savepoint within a transaction and then roll back all work done after the savepoint. Savepoints relax the atomicity property of transactions. A transaction with a savepoint is atomic in the sense that it appears to be a single unit outside the context of the transaction, but code operating within the transaction can preserve partial states.
Note:

Savepoints are supported for local transactions only. Specifying a savepoint within a global transaction causes a SQLException exception to be thrown.

Creating a Savepoint
You create a savepoint using the Connection.setSavepoint method, which returns a java.sql.Savepoint instance. A savepoint is either named or unnamed. You specify the name of a savepoint by supplying a string to the setSavepoint method. If you do not specify a name, then the savepoint is assigned an integer ID. You retrieve a name using the getSavepointName method. You retrieve an ID using the getSavepointId method.
Note:

Attempting to retrieve a name from an unnamed savepoint or attempting to retrieve an ID from a named savepoint throws a SQLException exception.

Rolling Back to a Savepoint

You roll back to a savepoint using the Connection.rollback(Savepoint svpt) method. If you try to roll back to a savepoint that has been released, then a SQLException exception is thrown.

JDBC Standards Support

3-3

Support for JDBC 3.0 Standard

Releasing a Savepoint
You remove a savepoint using the Connection.releaseSavepoint(Savepoint svpt) method.

Checking Savepoint Support

You query if savepoints are supported by your database by calling the oracle.jdbc.OracleDatabaseMetaData.supportsSavepoints method, which returns true if savepoints are available, false otherwise.

Savepoint Notes
When using savepoints, you must consider the following:

After a savepoint has been released, attempting to reference it in a rollback operation will cause a SQLException exception to be thrown. When a transaction is committed or rolled back, all savepoints created in that transaction are automatically released and become invalid. Rolling a transaction back to a savepoint automatically releases and makes invalid any savepoints created after the savepoint in question.

Retrieval of Auto-Generated Keys

Many database systems automatically generate a unique key field when a row is inserted. Oracle Database provides the same functionality with the help of sequences and triggers. JDBC 3.0 introduces the retrieval of auto-generated keys feature that enables you to retrieve such generated values. In JDBC 3.0, the following interfaces are enhanced to support the retrieval of auto-generated keys feature:

java.sql.DatabaseMetaData java.sql.Connection java.sql.Statement

These interfaces provide methods that support retrieval of auto-generated keys. However, this feature is supported only when INSERT statements are processed. Other data manipulation language (DML) statements are processed, but without retrieving auto-generated keys.

Note:

The Oracle server-side internal driver does not support the retrieval of auto-generated keys feature.

java.sql.Statement
If key columns are not explicitly indicated, then Oracle JDBC drivers cannot identify which columns need to be retrieved. When a column name or column index array is used, Oracle JDBC drivers can identify which columns contain auto-generated keys that you want to retrieve. However, when the Statement.RETURN_GENERATED_KEYS integer flag is used, Oracle JDBC drivers cannot identify these columns. When the integer flag is used to indicate that auto-generated keys are to be returned, the ROWID pseudo column is returned as key. The ROWID can be then fetched from the ResultSet object and can be used to retrieve other columns.

3-4 Oracle Database JDBC Developer's Guide and Reference

Support for JDBC 3.0 Standard

Sample Code
The following code illustrates retrieval of auto-generated keys:
/** SQL statements for creating an ORDERS table and a sequence for generating the * ORDER_ID. * * CREATE TABLE ORDERS (ORDER_ID NUMBER, CUSTOMER_ID NUMBER, ISBN NUMBER, * DESCRIPTION NCHAR(5)) * * CREATE SEQUENCE SEQ01 INCREMENT BY 1 START WITH 1000 */ ... String cols[] = {"ORDER_ID", "DESCRIPTION"}; // Create a PreparedStatement for inserting a row into the ORDERS table. OraclePreparedStatement pstmt = (OraclePreparedStatement) conn.prepareStatement("INSERT INTO ORDERS (ORDER_ID, CUSTOMER_ID, ISBN, DESCRIPTION) VALUES (SEQ01.NEXTVAL, 101, 966431502, ?)", cols); char c[] = {'a', '\u5185', 'b'}; String s = new String(c); pstmt.setNString(1, s); pstmt.executeUpdate(); ResultSet rset = pstmt.getGeneratedKeys(); ...

In the preceding example, a sequence, SEQ01, is created to generate values for the ORDER_ID column starting from 1000 and incrementing by 1 each time the sequence is processed to generate the next value. An OraclePreparedStatement object is created to insert a row in to the ORDERS table.

Limitations
Auto-generated keys are implemented using the DML returning clause. So, they are subjected to the following limitations:

You cannot combine auto-generated keys with batch update. You need to access the ResultSet object returned from getGeneratedKeys method by position only and no bind variable names should be used as columns in the ResultSet object.

JDBC 3.0 LOB Interface Methods

Table 32 and Table 33 show the conversions between Oracle proprietary methods and JDBC 3.0 standard methods.
Table 32 BLOB Method Equivalents JDBC 3.0 Standard Method setBytes(long pos, byte[] bytes) setBytes(long pos, byte[] bytes, int offset, int len) setBinaryStream(long pos) truncate(long len)

Oracle Proprietary Method putBytes(long pos, byte [] bytes) putBytes(long pos, byte [] bytes, int length) getBinaryOutputStream(long pos) trim (long len)

JDBC Standards Support

3-5

Support for JDBC 4.0 Standard

Table 33

CLOB Method Equivalents JDBC 3.0 Standard Method setString(long pos, String str) setString(long pos, String str, int offset, int len) setAsciiStream(long pos) setCharacterStream(long pos) truncate(long len)

Oracle Proprietary Method putString(long pos, String str) not applicable getAsciiOutputStream(long pos) getCharacterOutputStream(long pos) trim (long len)

Result Set Holdability

Result set holdability was introduced since JDBC 3.0. This feature enables applications to decide whether the ResultSet objects should be open or closed, when a commit operation is performed. The commit operation could be either implicit or explicit. Oracle Database supports only HOLD_CURSORS_OVER_COMMIT. Therefore, it is the default value for Oracle JDBC drivers. Any attempt to change holdability will throw a SQLException exception.

Support for JDBC 4.0 Standard

The JDBC 4.0 standard support is provided by JDK 1.6 and later versions. Oracle Database 11g Release 1 (11.1) JDBC drivers provide support for the JDBC 4.0 standard.
Note:

You need to have the ojdbc6*.jar in your classpath environment variable in order to have JDBC 4.0 standard support.

Some of the new features available in Oracle Database 11g Release 1 (11.1) JDBC drivers are the following:

Wrapper Pattern Support Enhanced Exception Hierarchy and SQLException The RowId Data Type LOB Creation National Language Character Set Support

This document provides only an overview of these new features. For detailed information about these features, see "Java 2 Platform, Standard Edition (JSE) 6.0 specification" at
http://java.sun.com/javase/6/docs/

Wrapper Pattern Support Wrapper pattern is a common coding pattern used in Java applications to provide extensions beyond the traditional JDBC API that are specific to a data source. You may need to use these extensions to access the resources that are wrapped as proxy class instances representing the actual resources. JDBC 4.0 introduces the Wrapper interface

3-6 Oracle Database JDBC Developer's Guide and Reference

Support for JDBC 4.0 Standard

that describes a standard mechanism to access these wrapped resources represented by their proxy, to permit direct access to the resource delegates. The Wrapper interface provides the following two methods:

public boolean isWrapperFor(Class<?> iface) throws SQLException; public <T> T unwrap(Class<T> iface) throws SQLException;

The other JDBC 4.0 interfaces, except those that represent SQL data, all implement this interface. These include Connection, Statement and its subtypes, ResultSet, and the metadata interfaces.
See Also:

http://java.sun.com/javase/6/docs/api/java/sql/Wrapp er.html Enhanced Exception Hierarchy and SQLException JDBC 3.0 defines only a single exception, SQLException. However, there are large categories of errors and it is useful to distinguish them. This feature provides subclasses of the SQLException class to identify the different categories of errors. The primary distinction is between permanent errors and transient errors. Permanent errors are a result of the correct operation of the system and will always occur. Transient errors are the result of failures, including timeouts, of some part of the system and may not reoccur. New exceptions have been added to represent transient and permanent errors and the different categories of these errors. Also, the SQLException class and its subclasses have been enhanced to provide support for the J2SE chained exception functionality. The RowId Data Type JDBC 4.0 provides the java.sql.RowId data type to represent SQL ROWID values. You can retrieve a RowId value using the getter methods defined in the ResultSet and CallableStatement interfaces. You can also use a RowId value in a parameterized PreparedStatement to set a parameter with a RowId object or in an updatable result set to update a column with a specific RowId value. A RowId object is valid until the identified row is not deleted. A RowId object may also be valid for the following:

The duration of the transaction in which it is created The duration of the session in which it is created An undefined duration where by it is valid forever

The lifetime of the RowId object can be determined by calling the DatabaseMetaData.getRowIdLifetime method. LOB Creation In JDBC 4.0, the Connection interface has been enhanced to provide support for the creation of BLOB, CLOB, and NCLOB objects. The interface provides the createBlob, createClob, and createNClob methods that enable you to create Blob, Clob, and NClob objects. The created large objects (LOBs) do not contain any data. You can add data in these objects by calling the appropriate setXXX methods. To retrieve the data from these

JDBC Standards Support

3-7

Support for JDBC 4.0 Standard

objects, you can call the getBlob, getClob, and getNClob methods defined in the ResultSet and CallableStatement interfaces. You can either retrieve the entire content or a part of the content from these objects. The following code snippet illustrates how to retrieve 100 bytes of data from a BLOB object starting at offset 200:
... Connection con = DriverManager.getConnection(url, props); Blob aBlob = con.createBlob(); // Add data to the BLOB object. ... // Retrieve part of the data from the BLOB object. InputStream is = aBlob.getBinaryStream(200, 100); ...

You can also pass LOBs as input parameters to a PreparedStatement object using the setBlob, setClob, and setNClob methods. You can use the updateBlob, updateClob, and updateNClob methods to update a column value in an updatable result set. LOBs remain valid for at least the duration of the transaction in which they are created. This may result in unwarranted use of memory during a long running transaction. You can release LOBs by calling their free method, as follows:
... Clob aClob = con.createClob(); int numWritten = aClob.setString(1, val); aClob.free(); ...

National Language Character Set Support JDBC 4.0 introduces the NCHAR, NVARCHAR, LONGNVARCHAR, and NCLOB JDBC types to access the national character set types. These types are similar to the CHAR, VARCHAR, LONGVARCHAR, and CLOB types, except that the values are encoded using the national character set.

3-8 Oracle Database JDBC Developer's Guide and Reference

4
Oracle Extensions
Oracle provides Java classes and interfaces that extend the Java Database Connectivity (JDBC) standard implementation, enabling you to access and manipulate Oracle data types and use Oracle performance extensions. Compared to standard JDBC, the Oracle extensions offer greater flexibility in manipulating the data. This chapter provides an overview of the classes and interfaces provided by Oracle that extend the JDBC standard implementation. It also describes some of the key support features of the extensions. This chapter contains the following sections:

Overview of Oracle Extensions Features of the Oracle Extensions Oracle JDBC Packages Oracle Character Data Types Support Additional Oracle Type Extensions DML Returning Accessing PL/SQL Index-by Tables
Note:

This chapter focuses on type extensions, as opposed to performance extensions, which are discussed in detail in Chapter 23, "Performance Extensions".

Overview of Oracle Extensions

Beyond standard features, Oracle JDBC drivers provide Oracle-specific type extensions and performance extensions. These extensions are provided through the following Java packages:

oracle.sql Provides classes that represent SQL data in Oracle format

oracle.jdbc Provides interfaces to support database access and updates in Oracle type formats
See Also:

"Oracle JDBC Packages" on page 4-5

Oracle Extensions

4-1

Features of the Oracle Extensions

Features of the Oracle Extensions

The Oracle extensions to JDBC include a number of features that enhance your ability to work with Oracle Databases. These include the following:

Database Management Using JDBC Support for Oracle Data Types Support for Oracle Objects Support for Schema Naming DML Returning Accessing PL/SQL Index-by Tables

Database Management Using JDBC

Oracle Database 11g Release 1 (11.1) introduces new JDBC methods, startup and shutdown, in the oracle.jdbc.OracleConnection interface that enable you to start up and shut down an Oracle Database instance. You also have support for the Database Change Notification feature of Oracle Database. These new features have been discussed in details in "Database Management".

Support for Oracle Data Types

A key feature of the Oracle JDBC extensions is the type support in the oracle.sql package. This package includes classes that map to all the Oracle SQL data types, acting as containers for raw SQL data. This functionality provides two significant advantages in manipulating SQL data:

Accessing data directly in SQL format is sometimes more efficient than first converting it to Java format. Performing mathematical manipulations of the data directly in SQL format avoids the loss of precision that can occur in converting between SQL and Java formats. But this type of conversion reduces the performance of the application.

Once manipulations are complete and it is time to display the information, each of the oracle.sql.* data type support classes has all the necessary methods to convert data to appropriate Java formats.
See Also:

Package oracle.sql on page 4-5 "Oracle Character Data Types Support" on page 4-10 "Additional Oracle Type Extensions" on page 4-13

Oracle recommends you to use the standard JDBC types unless there is a specific and critical need to use the oracle.sql types. Oracle JDBC drivers are designed to optimize the use of the JDBC standard types. This is very important if you are using character data. Always use java.lang.String for character data instead of oracle.sql.CHAR. CHAR is provided only for backward compatibility.
Note:

4-2 Oracle Database JDBC Developer's Guide and Reference

Features of the Oracle Extensions

Support for Oracle Objects

Oracle JDBC supports the use of structured objects in the database, where an object data type is a user-defined type with nested attributes. For example, a user application could define an Employee object type, where each Employee object has a firstname attribute (character string), a lastname attribute (character string), and an employeenumber attribute (integer). Oracle JDBC supports Oracle object data types. When you work with Oracle object data types in a Java application, you must consider the following:

How to map between Oracle object data types and Java classes How to store Oracle object attributes in corresponding Java objects How to convert attribute data between SQL and Java formats How to access data

Oracle objects can be mapped either to the weak java.sql.Struct type or to strongly typed customized classes. These strong types are referred to as custom Java classes, which must implement either the standard java.sql.SQLData interface or the Oracle extension oracle.sql.ORAData interface. Each interface specifies methods to convert data between SQL and Java.
Note: The ORAData interface has replaced the CustomDatum interface. The latter interface is desupported in Oracle Database release 11.1.

Oracle recommends the use of the Oracle JPublisher utility to create custom Java classes to correspond to your Oracle objects. Oracle JPublisher performs this task seamlessly with command-line options and can generate either SQLData or ORAData interface implementations. For SQLData interface implementations, a type map defines the correspondence between Oracle object data types and Java classes. Type maps are objects that specify which Java class corresponds to each Oracle object data type. Oracle JDBC uses these type maps to determine which Java class to instantiate and populate when it retrieves Oracle object data from a result set.
Note:

Oracle recommends using the ORAData interface, instead of the SQLData interface, in situations where portability is not a concern. The ORAData interface works more easily and flexibly in conjunction with other features of the Oracle platform offerings using Java.

JPublisher automatically defines getXXX methods of the custom Java classes, which retrieve data into your Java application.
See Also:

Chapter 13, "Working with Oracle Object Types" Oracle Database JPublisher User's Guide.

Oracle Extensions

4-3

Features of the Oracle Extensions

Support for Schema Naming

Oracle object data type classes have the ability to accept and return fully qualified schema names. A fully qualified schema name has this syntax:
{[schema_name].}[sql_type_name]

Where, schema_name is the name of the schema and sql_type_name is the SQL type name of the object. schema_name and sql_type_name are separated by a period (.). To specify an object type in JDBC, use its fully qualified name. It is not necessary to enter a schema name if the type name is in the current naming space, that is, the current schema. Schema naming follows these rules:

Both the schema name and the type name may or may not be within quotation marks. However, if the SQL type name has a period in it, such as CORPORATE.EMPLOYEE, the type name must be quoted. The JDBC driver looks for the first period in the object name that is not within quotation marks and uses the string before the period as the schema name and the string following the period as the type name. If no period is found, then the JDBC driver takes the current schema as default. That is, you can specify only the type name, without indicating a schema, instead of specifying the fully qualified name if the object type name belongs to the current schema. This also explains why you must put the type name within quotation marks if the type name has a dot in it. For example, assume that user Scott creates a type called person.address and then wants to use it in his session. Scott may want to skip the schema name and pass in person.address to the JDBC driver. In this case, if person.address is not within quotation marks, then the period will be detected and the JDBC driver will mistakenly interpret person as the schema name and address as the type name.

JDBC passes the object type name string to the database unchanged. That is, the JDBC driver will not change the character case even if the object type name is within quotation marks. For example, if Scott.PersonType is passed to the JDBC driver as an object type name, then the JDBC driver will pass the string to the database unchanged. As another example, if there is white space between characters in the type name string, then the JDBC driver will not remove the white space.

DML Returning
Oracle Database supports the use of the RETURNING clause with data manipulation language (DML) statements. This enables you to combine two SQL statements into one. Both the Oracle JDBC Oracle Call Interface (OCI) driver and the Oracle JDBC Thin driver support DML returning.
See Also:

"DML Returning" on page 4-26

Accessing PL/SQL Index-by Tables

Oracle JDBC drivers enable JDBC applications to make PL/SQL calls with index-by table parameters. Oracle JDBC drivers support PL/SQL index-by tables of scalar data types
Note:

Index-by tables of PL/SQL records are not supported.

4-4 Oracle Database JDBC Developer's Guide and Reference

Oracle JDBC Packages

See Also:

"Accessing PL/SQL Index-by Tables" on page 4-28

Oracle JDBC Packages

This section describes the following Java packages, which support the Oracle JDBC extensions:

Package oracle.sql Package oracle.jdbc

Package oracle.sql
The oracle.sql package supports direct access to data in SQL format. This package consists primarily of classes that provide Java mappings to SQL data types and their support classes. Essentially, the classes act as Java containers for SQL data. Each of the oracle.sql.* data type classes extends oracle.sql.Datum, a superclass that encapsulates functionality common to all the data types. Some of the classes are for JDBC 2.0-compliant data types. These classes, as Table 41 indicates, implement standard JDBC 2.0 interfaces in the java.sql package, as well as extending the oracle.sql.Datum class. Oracle recommends the use of standard JDBC types or Java types whenever possible. The types in the package oracle.sql.* are provided primarily for backward compatibility or for support of a few Oracle specific features such as OPAQUE, OraData, TIMESTAMPTZ, and so on.
Note:

Classes of the oracle.sql Package Table 41 lists the oracle.sql data type classes and their corresponding Oracle SQL types.
Table 41 Java Class oracle.sql.STRUCT oracle.sql.REF oracle.sql.ARRAY oracle.sql.BLOB oracle.sql.CLOB Oracle Data Type Classes Oracle SQL Types and Interfaces Implemented STRUCT (objects) implements java.sql.Struct REF (object references) implements java.sql.Ref VARRAY or nested table (collections) implements java.sql.Array BLOB (binary large objects) implements java.sql.Blob SQL CLOB (character large objects) and globalization support NCLOB data types both implement java.sql.Clob NCLOB implements java.sql.NClob BFILE (external files) CHAR, NCHAR, VARCHAR2, NVARCHAR2 DATE TIMESTAMP TIMESTAMP WITH TIME ZONE

oracle.sql.NCLOB oracle.sql.BFILE oracle.sql.CHAR oracle.sql.DATE oracle.sql.TIMESTAMP oracle.sql.TIMESTAMPTZ

Oracle Extensions

4-5

Oracle JDBC Packages

Table 41 (Cont.) Oracle Data Type Classes Java Class oracle.sql.TIMESTAMPLTZ oracle.sql.NUMBER oracle.sql.RAW oracle.sql.ROWID oracle.sql.OPAQUE oracle.sql.ANYDATA Oracle SQL Types and Interfaces Implemented TIMESTAMP WITH LOCAL TIME ZONE NUMBER RAW ROWID (row identifiers) implements java.sql.RowId OPAQUE ANYDATA

Note: The LONG and LONG RAW SQL types and REF CURSOR type category have no oracle.sql.* classes. Use standard JDBC functionality for these types. For example, retrieve LONG or LONG RAW data as input streams using the standard JDBC result set and callable statement methods getBinaryStream and getCharacterStream. Use the getCursor method for REF CURSOR types.

In addition to the data type classes, the oracle.sql package includes the following support classes and interfaces, primarily for use with objects and collections:

oracle.sql.ArrayDescriptor This class is used in constructing oracle.sql.ARRAY objects. It describes the SQL type of the array.

oracle.sql.StructDescriptor This class is used in constructing oracle.sql.STRUCT objects, which you can use as a default mapping to Oracle objects in the database.

oracle.sql.ORAData and oracle.sql.ORADataFactory These interfaces are used in Java classes implementing the Oracle ORAData scenario of Oracle object support.

oracle.sql.OpaqueDescriptor This class is used to obtain the metadata for an instance of the oracle.sql.OPAQUE class.

oracle.sql.TypeDescriptor This class is used to represent transient and persistent SQL types in Java.

General oracle.sql.* Data Type Support Each of the Oracle data type classes provides, among other things, the following:

Data storage as Java byte arrays for SQL data A getBytes() method, which returns the SQL data as a byte array A toJdbc() method that converts the data into an object of a corresponding Java class as defined in the JDBC specification The JDBC driver does not convert Oracle-specific data types that are not part of the JDBC specification, such as BFILE. The driver returns the object in the corresponding oracle.sql.* format.

4-6 Oracle Database JDBC Developer's Guide and Reference

Oracle JDBC Packages

Appropriate xxxValue methods to convert SQL data to Java type. For example, stringValue, intValue, booleanValue, dateValue, and bigDecimalValue Additional conversion methods, getXXX and setXXX, as appropriate, for the functionality of the data type, such as methods in the large object (LOB) classes that get the data as a stream and methods in the REF class that get and set object data through the object reference.

Overview of Class oracle.sql.STRUCT For any given Oracle object type, it is usually desirable to define a custom mapping between SQL and Java. For example, if you use a SQLData custom Java class, then the mapping must be defined in a type map. If you choose not to define a mapping, however, then data from the object type will be materialized in Java in an instance of the oracle.sql.STRUCT class. The STRUCT class implements the standard JDBC 2.0 java.sql.Struct interface and extends the oracle.sql.Datum class. A STRUCT object is a Java representation of the raw bytes of an Oracle object. It contains the SQL type name of the Oracle object and an array of oracle.sql.Datum objects that hold the attribute values in SQL format. If you want to create a STRUCT object, then use the createStruct method of the oracle.jdbc.OracleConnection interface. The signature of this factory method for creating STRUCT objects is as follows:
Struct createStruct (String typeName, Object[] attributes) throws SQLException

The parameters in this signature are as follows:

The typeName parameter is the SQL type name of the SQL structured type to which the STRUCT object maps. The typeName is the name of a user-defined type that has been defined for this database. It is the value returned by the Struct.getSQLTypeName method. The attributes parameter specifies the attributes that populate the returned object.

You can materialize attributes of a STRUCT object as oracle.sql.Datum[] objects, if you use the getOracleAttributes method, or as java.lang.Object[] objects, if you use the getAttributes method. Materializing the attributes as oracle.sql.* objects gives you the following advantages of the oracle.sql.* format:

Materializing oracle.sql.STRUCT data in oracle.sql.* format completely preserves data by maintaining it in SQL format. No translation is performed. This is useful if you want to access data but not necessarily display it. It allows complete flexibility in how your Java application unpacks data.

Oracle Extensions

4-7

Oracle JDBC Packages

Note:

Elements of the array, although of the generic Datum type, actually contain data associated with the relevant oracle.sql.* type appropriate for the given attribute. You can cast the element to the appropriate oracle.sql.* type as desired. For example, a CHAR data attribute within the STRUCT is materialized as oracle.sql.Datum. To use it as CHAR data, you must cast it to oracle.sql.CHAR. Nested objects in the values array of a STRUCT object are materialized by the JDBC driver as instances of STRUCT.

Overview of Class oracle.sql.REF The oracle.sql.REF class is the generic class that supports Oracle object references. This class, as with all oracle.sql.* data type classes, is a subclass of the oracle.sql.Datum class. It implements the standard JDBC 2.0 java.sql.Ref interface. The REF class has methods to retrieve and pass object references. However, selecting an object reference retrieves only a pointer to an object. This does not materialize the object itself. But the REF class also includes methods to retrieve and pass the object data. You cannot create REF objects in your JDBC application. You can only retrieve existing REF objects from the database.
See Also:

Chapter 15, "Using Oracle Object References".

Overview of Class oracle.sql.ARRAY The oracle.sql.ARRAY class supports Oracle collections, either VARRAYs or nested tables. If you select either a VARRAY or a nested table from the database, then the JDBC driver materializes it as an object of the ARRAY class. The structure of the data is equivalent in either case. The oracle.sql.ARRAY class extends the oracle.sql.Datum class and implements the standard JDBC 2.0 java.sql.Array interface. You can use the setARRAY method of the OraclePreparedStatement or OracleCallableStatement interface to pass an ARRAY as an input parameter to a prepared statement. Similarly, you can use the createARRAY method of the OracleConnection interface to create an ARRAY object to pass it to a prepared statement or callable statement, perhaps to insert into the database.
See Also:

"Overview of Collection Functionality" on page 16-3

Overview of Classes oracle.sql.BLOB, oracle.sql.CLOB, oracle.sql.BFILE Binary large objects (BLOBs), character large objects (CLOBs), and binary files (BFILEs) are for data items that are too large to store directly in a database table. Instead, the database table stores a locator that points to the location of the actual data. The oracle.sql package supports these data types in several ways:

BLOBs point to large unstructured binary data items and are supported by the oracle.sql.BLOB class. CLOBs point to large character data items and are supported by the oracle.sql.CLOB class.

4-8 Oracle Database JDBC Developer's Guide and Reference

Oracle JDBC Packages

BFILEs point to the content of external files (operating system files) and are supported by the oracle.sql.BFILE class. BFiles are read-only.

You can select a BLOB, CLOB, or BFILE locator from the database using a standard SELECT statement. However, you receive only the locator, and not the data. Additional steps are necessary to retrieve the data.
See Also:

Chapter 14, "Working with LOBs and BFILEs".

Classes oracle.sql.DATE, oracle.sql.NUMBER, and oracle.sql.RAW These classes map to primitive SQL data types, which are a part of standard JDBC, and supply conversions to and from the corresponding JDBC Java types. Because Java Double and Float NaN values do not have an equivalent Oracle NUMBER representation, a NullPointerException is thrown whenever a Double.NaN value or a Float.NaN value is converted into an Oracle NUMBER using the oracle.sql.NUMBER class. For instance, the following code throws a NullPointerException:
oracle.sql.NUMBER n = new oracle.sql.NUMBER(Double.NaN); System.out.println(n.doubleValue()); // throws NullPointerException

Classes oracle.sql.TIMESTAMP, oracle.sql.TIMESTAMPTZ, and oracle.sql.TIMESTAMPLTZ The JDBC drivers support the following date/time data types:

TIMESTAMP (TIMESTAMP) TIMESTAMP WITH TIME ZONE (TIMESTAMPTZ) TIMESTAMP WITH LOCAL TIME ZONE (TIMESTAMPLTZ)

The JDBC drivers allow conversions between DATE and date/time data types. For example, you can access a TIMESTAMP WITH TIME ZONE column as a DATE value. The JDBC drivers support the most popular time zone names used in the industry as well as most of the time zone names defined in the JDK. Time zones are specified by using the java.util.Calendar class.
Note:

Do not use TimeZone.getTimeZone to create time zone objects. The Oracle time zone data types support more time zone names than does the JDK.

The following code shows how the TimeZone and Calendar objects are created for US_PACIFIC, which is a time zone name not defined in the JDK:
TimeZone tz = TimeZone.getDefault(); tz.setID("US_PACIFIC"); GregorianCalendar gcal = new GregorianCalendar(tz);

The following Java classes represent the SQL date/time types:

oracle.sql.TIMESTAMP oracle.sql.TIMESTAMPTZ oracle.sql.TIMESTAMPLTZ

Before accessing TIMESTAMP WITH LOCAL TIME ZONE data, call the OracleConnection.setSessionTimeZone(String regionName) method to

Oracle Extensions

4-9

Oracle Character Data Types Support

set the session time zone. When this method is called, the JDBC driver sets the session time zone of the connection and saves the session time zone so that any TIMESTAMP WITH LOCAL TIME ZONE data accessed through JDBC can be adjusted using the session time zone. Class oracle.sql.OPAQUE The oracle.sql.OPAQUE class gives you the name and characteristics of the OPAQUE type and any attributes. The OPAQUE type provides access only to the uninterrupted bytes of the instance.
Note:

There is minimal support for the OPAQUE type.

Package oracle.jdbc
The interfaces of the oracle.jdbc package define the Oracle extensions to the interfaces in java.sql. These extensions provide access to Oracle SQL-format data and other Oracle-specific functionality, including Oracle performance enhancements.
See Also:

"The oracle.jdbc Package" on page 4-19

Oracle Character Data Types Support

Oracle character data types include the SQL CHAR and NCHAR data types. The following sections describe how these data types can be accessed using the oracle.sql.* classes:

SQL CHAR Data Types SQL NCHAR Data Types Class oracle.sql.CHAR

SQL CHAR Data Types

The SQL CHAR data types include CHAR, VARCHAR2, and CLOB. These data types let you store character data in the database character set encoding scheme. The character set of the database is established when you create the database.

SQL NCHAR Data Types

The SQL NCHAR data types were created for Globalization Support. The SQL NCHAR data types include NCHAR, NVARCHAR2, and NCLOB. These data types allow you to store Unicode data in the database NCHAR character set encoding. The NCHAR character set, which never changes, is established when you create the database.
Note:

Because the UnicodeStream class is deprecated in favor of the CharacterStream class, the setUnicodeStream and getUnicodeStream methods are not supported for NCHAR data type access. Use the setCharacterStream method and the getCharacterStream method if you want to use stream access.

The usage of SQL NCHAR data types is similar to that of the SQL CHAR data types. JDBC uses the same classes and methods to access SQL NCHAR data types that are used for the corresponding SQL CHAR data types. Therefore, there are no separate,

4-10 Oracle Database JDBC Developer's Guide and Reference

Oracle Character Data Types Support

corresponding classes defined in the oracle.sql package for SQL NCHAR data types. Similarly, there is no separate, corresponding constant defined in the oracle.jdbc.OracleTypes class for SQL NCHAR data types. The following code shows how to access SQL NCHAR data:
// // Table TEST has the following columns: // - NUMBER // - NVARCHAR2 // - NCHAR // oracle.jdbc.OraclePreparedStatement pstmt = (oracle.jdbc.OraclePreparedStatement) conn.prepareStatement("insert into TEST values(?, ?, ?)"); // // oracle.jdbc.OraclePreparedStatement.FORM_NCHAR should be used for all NCHAR, // NVARCHAR2 and NCLOB data types. // pstmt.setInt(1, 1); // NUMBER column pstmt.setNString(2, myUnicodeString1); // NVARCHAR2 column pstmt.setNString(3, myUnicodeString2); // NCHAR column pstmt.execute();

Class oracle.sql.CHAR
The oracle.sql.CHAR class is used by Oracle JDBC in handling and converting character data. This class provides the Globalization Support functionality to convert character data. This class has two key attributes: Globalization Support character set and the character data. The Globalization Support character set defines the encoding of the character data. It is a parameter that is always passed when a CHAR object is constructed. Without the Globalization Support character set information, the data bytes in the CHAR object are meaningless. The oracle.sql.CHAR class is used for both SQL CHAR and SQL NCHAR data types.
Note: In versions of Oracle JDBC drivers prior to 10g release 1 (10.1), there were performance advantages to using the oracle.SQL.CHAR. Starting from Oracle Database 10g, there are no longer any such advantages. In fact, optimum performance is achieved using the java.lang.String. All Oracle JDBC drivers handle all character data in the Java UCS2 character set. Using the oracle.sql.CHAR does not prevent conversions between the database character set and UCS2 character set.

The only remaining use of the oracle.sql.CHAR class is to handle character data in the form of raw bytes encoded in an Oracle Globalization Support character set. All character data retrieved from Oracle Database should be accessed using the java.lang.String class. When processing byte data from another source, you can use an oracle.sql.CHAR to convert the bytes to java.lang.String. To convert an oracle.sql.CHAR, you must provide the data bytes and an oracle.sql.CharacterSet instance that represents the Globalization Support character set used to encode the data bytes.

Oracle Extensions 4-11

Oracle Character Data Types Support

The CHAR objects that are Oracle object attributes are returned in the database character set. JDBC application code rarely needs to construct CHAR objects directly, because the JDBC driver automatically creates CHAR objects as needed. To construct a CHAR object, you must provide character set information to the CHAR object by way of an instance of the CharacterSet class. Each instance of this class represents one of the Globalization Support character sets that Oracle supports. A CharacterSet instance encapsulates methods and attributes of the character set, mainly involving functionality to convert to or from other character sets. Constructing an oracle.sql.CHAR Object Follow these general steps to construct a CHAR object:
1.

Create a CharacterSet object by calling the static CharacterSet.make method. This method is a factory for the character set instance. The make method takes an integer as input, which corresponds to a character set ID that Oracle supports. For example:
int oracleId = CharacterSet.JA16SJIS_CHARSET; // this is character set ID, // 832 ... CharacterSet mycharset = CharacterSet.make(oracleId);

Each character set that Oracle supports has a unique, predefined Oracle ID.
2.

Construct a CHAR object. Pass a string, or the bytes that represent the string, to the factory method along with the CharacterSet object that indicates how to interpret the bytes based on the character set. For example:
String mystring = "teststring"; ... CHAR mychar = new CHAR(teststring, mycharset);

There are multiple factory methods for CHAR, which can take a String, a byte array, or an object as input along with the CharacterSet object. In the case of a String, the string is converted to the character set indicated by the CharacterSet object before being placed into the CHAR object.
Note:

The CharacterSet object cannot be a null value. The CharacterSet class is an abstract class, therefore it has no constructor. The only way to create instances is to use the make method. The server recognizes the special value CharacterSet.DEFAULT_CHARSET as the database character set. For the client, this value is not meaningful. Oracle does not intend or recommend that users extend the CharacterSet class.

4-12 Oracle Database JDBC Developer's Guide and Reference

Additional Oracle Type Extensions

oracle.sql.CHAR Conversion Methods The CHAR class provides the following methods for translating character data to strings:

getString This method converts the sequence of characters represented by the CHAR object to a string, returning a Java String object. If you enter an invalid OracleID, then the character set will not be recognized and the getString method will throw a SQLException exception.

toString This method is identical to the getString method. But if you enter an invalid OracleID, then the character set will not be recognized and the toString method will return a hexadecimal representation of the CHAR data and will not throw a SQLException exception.

getStringWithReplacement This method is identical to the getString method, except a default replacement character replaces characters that have no unicode representation in the CHAR object character set. This default character varies from character set to character set, but is often a question mark (?).

The database server and the client, or application running on the client, can use different character sets. When you use the methods of the CHAR class to transfer data between the server and the client, the JDBC drivers must convert the data from the server character set to the client character set or vice versa. To convert the data, the drivers use Globalization Support.
See Also:

Chapter 19, "Globalization Support"

Additional Oracle Type Extensions

Oracle JDBC drivers support the Oracle-specific BFILE and ROWID data types and REF CURSOR types, which are not part of the standard JDBC specification. This section describes the ROWID and REF CURSOR type extensions. The ROWID is supported as a Java string, and REF CURSOR types are supported as JDBC result sets. This section covers the following topics:

Oracle ROWID Type Oracle REF CURSOR Type Category Oracle BINARY_FLOAT and BINARY_DOUBLE Types Oracle SYS.ANYTYPE and SYS.ANYDATA Types The oracle.jdbc Package

Oracle ROWID Type

A ROWID is an identification tag unique for each row of an Oracle Database table. The ROWID can be thought of as a virtual column, containing the ID for each row. The oracle.sql.ROWID class is supplied as a container for ROWID SQL data type. ROWIDs provide functionality similar to the getCursorName method specified in the java.sql.ResultSet interface and the setCursorName method specified in the java.sql.Statement interface.
Oracle Extensions 4-13

Additional Oracle Type Extensions

If you include the ROWID pseudo-column in a query, then you can retrieve the ROWIDs with the result set getString method. You can also bind a ROWID to a PreparedStatement parameter with the setString method. This enables in-place updating, as in the example that follows.
Note: The oracle.sql.ROWID class replaces oracle.jdbc.driver.ROWID, which was used in previous releases of Oracle JDBC. But, use the former class only when using J2SE 1.5. For JSE 6, use the java.sql.RowId interface instead.

Example The following example shows how to access and manipulate ROWID data:
Statement stmt = conn.createStatement(); // Query the employee names with "FOR UPDATE" to lock the rows. // Select the ROWID to identify the rows to be updated. ResultSet rset = stmt.executeQuery ("SELECT ename, rowid FROM emp FOR UPDATE"); // Prepare a statement to update the ENAME column at a given ROWID PreparedStatement pstmt = conn.prepareStatement ("UPDATE emp SET ename = ? WHERE rowid = ?"); // Loop through the results of the query while (rset.next ()) { String ename = rset.getString (1); oracle.sql.ROWID rowid = rset.getROWID (2); // Get the ROWID as a String pstmt.setString (1, ename.toLowerCase ()); pstmt.setROWID (2, rowid); // Pass ROWID to the update statement pstmt.executeUpdate (); // Do the update }

Oracle REF CURSOR Type Category

A cursor variable holds the memory location of a query work area, rather than the contents of the area. Declaring a cursor variable creates a pointer. In SQL, a pointer has the data type REF x, where REF is short for REFERENCE and x represents the entity being referenced. A REF CURSOR, then, identifies a reference to a cursor variable. Because many cursor variables might exist to point to many work areas, REF CURSOR can be thought of as a category or data type specifier that identifies many different types of cursor variables.
Note:

REF CURSOR instances are not scrollable.

To create a cursor variable, begin by identifying a type that belongs to the REF CURSOR category. For example:
DECLARE TYPE DeptCursorTyp IS REF CURSOR

Then, create the cursor variable by declaring it to be of the type DeptCursorTyp:

4-14 Oracle Database JDBC Developer's Guide and Reference

Additional Oracle Type Extensions

dept_cv DeptCursorTyp ...

- - declare cursor variable

REF CURSOR, then, is a category of data types, rather than a particular data type. Stored procedures can return cursor variables of the REF CURSOR category. This output is equivalent to a database cursor or a JDBC result set. A REF CURSOR essentially encapsulates the results of a query. In JDBC, a REF CURSOR is materialized as a ResultSet object and can be accessed as follows:
1.

Use a JDBC callable statement to call a stored procedure. It must be a callable statement, as opposed to a prepared statement, because there is an output parameter. The stored procedure returns a REF CURSOR. The Java application casts the callable statement to an Oracle callable statement and uses the getCursor method of the OracleCallableStatement class to materialize the REF CURSOR as a JDBC ResultSet object. The result set is processed as requested.
Important:

2. 3.

4.

The cursor associated with a REF CURSOR is closed whenever the statement object that produced the REF CURSOR is closed. Unlike in past releases, the cursor associated with a REF CURSOR is not closed when the result set object in which the REF CURSOR was materialized is closed.

Example This example shows how to access REF CURSOR data.

import oracle.jdbc.*; ... CallableStatement cstmt; ResultSet cursor; // Use a PL/SQL block to open the cursor cstmt = conn.prepareCall ("begin open ? for select ename from emp; end;"); cstmt.registerOutParameter(1, OracleTypes.CURSOR); cstmt.execute(); cursor = ((OracleCallableStatement)cstmt).getCursor(1); // Use the cursor like a standard ResultSet while (cursor.next ()) {System.out.println (cursor.getString(1));}

In the preceding example:

A CallableStatement object is created by using the prepareCall method of the connection class. The callable statement implements a PL/SQL procedure that returns a REF CURSOR.

Oracle Extensions 4-15

Additional Oracle Type Extensions

As always, the output parameter of the callable statement must be registered to define its type. Use the type code OracleTypes.CURSOR for a REF CURSOR. The callable statement is run, returning the REF CURSOR. The CallableStatement object is cast to OracleCallableStatement to use the getCursor method, which is an Oracle extension to the standard JDBC API, and returns the REF CURSOR into a ResultSet object.

Oracle BINARY_FLOAT and BINARY_DOUBLE Types

The Oracle BINARY_FLOAT and BINARY_DOUBLE types are used to store IEEE 574 float and double data. These correspond to the Java float and double scalar types with the exception of negative zero and NaN.
See Also:

Oracle Database SQL Language Reference

If you include a BINARY_DOUBLE column in a query, then the data is retrieved from the database in the binary format. Also, the getDouble method will return the data in the binary format. In contrast, for a NUMBER data type column, the number bits are returned and converted to the Java double data type.
Note: The Oracle representation for the SQL FLOAT, DOUBLE PRECISION, and REAL data types use the Oracle NUMBER representation. The BINARY_FLOAT and BINARY_DOUBLE data types can be regarded as proprietary types.

A call to the JDBC standard setDouble(int, double) method of the PreparedStatement interface converts the Java double argument to Oracle NUMBER style bits and send them to the database. In contrast, the setBinaryDouble(int, double) method of the oracle.jdbc.OraclePreparedStatement interface converts the data to the internal binary bits and sends them to the database. You must ensure that the data format used matches the type of the target parameter of the PreparedStatement interface. This will result in correct data and least use of CPU. If you use setBinaryDouble for a NUMBER parameter, then the binary bits are sent to the server and converted to NUMBER format. The data will be correct, but server CPU load will be increased. If you use setDouble for a BINARY_DOUBLE parameter, then the data will first be converted to NUMBER bits on the client and sent to the server, where it will be converted back to binary format. This will increase the CPU load on both client and server and can result in data corruption as well. The SetFloatAndDoubleUseBinary connection property when set to true causes the JDBC standard APIs, setFloat(int, float), setDouble(int, double), and all the variations, to send internal binary bits instead of NUBMER bits.
Note:

Although this section largely discusses BINARY_DOUBLE, the same is true for BINARY_FLOAT.

Oracle SYS.ANYTYPE and SYS.ANYDATA Types

Oracle Database 11g Release 1 (11.1) provides a Java interface to access the SYS.ANYTYPE and SYS.ANYDATA Oracle types.

4-16 Oracle Database JDBC Developer's Guide and Reference

Additional Oracle Type Extensions

See Also: For information about these Oracle types, refer Oracle Database PL/SQL Packages and Types Reference

An instance of the SYS.ANYTYPE type contains a type description of any SQL type, persistent or transient, named or unnamed, including object types and collection types. You can use the oracle.sql.TypeDescriptor class to access the SYS.ANYTYPE type. An ANYTYPE instance can be retrieved from a PL/SQL procedure or a SQL SELECT statement where SYS.ANYTYPE is used as a column type. To retrieve an ANYTYPE instance from the database, use the getObject method. This method returns an instance of the TypeDescriptor. The retrieved ANYTYPE instance could be any of the following:

Transient object type Transient predefined type Persistent object type Persistent predefined type

Example 41 Code Snippet for Accessing SYS.ANYTYPE Type

The following code snippet illustrates how to retrieve an instance on ANYTYPE from the database:
... ResultSet rs = stmt.executeQuery("select anytype_column from my_table"); TypeDescriptor td = (TypeDescriptor)rs.getObject(1); short typeCode = td.getInternalTypeCode(); if(typeCode == TypeDescriptor.TYPECODE_OBJECT) { // check if it's a transient type if(td.isTransientType()) { AttributeDescriptor[] attributes = ((StructDescriptor)td).getAttributesDescriptor(); for(int i=0; i<attributes.length; i++) System.out.println(attributes[i].getAttributeName()); } else { System.out.println(td.getTypeName()); } } ... Example 42 Creating a Transient Object Type Through PL/SQL and Retrieving Through JDBC

This example provides a code snippet illustrating how to retrieve a transient object type through JDBC.
... OracleCallableStatement cstmt = (OracleCallableStatement)conn.prepareCall ("BEGIN ? := transient_obj_type (); END;"); cstmt.registerOutParameter(1,OracleTypes.OPAQUE,"SYS.ANYTYPE"); cstmt.execute(); TypeDescriptor obj = (TypeDescriptor)cstmt.getObject(1); if(!obj.isTransient()) System.out.println("This must be a JDBC bug"); cstmt.close();

Oracle Extensions 4-17

Additional Oracle Type Extensions

return obj; ...

Example 43 Calling a PL/SQL Stored Procedure That Takes an ANYTPE as IN Parameter

The following code snippet illustrates how to call a PL/SQL stored procedure that takes an ANYTYPE as IN parameter:
... CallableStatement cstmt = conn.prepareCall("BEGIN ? := dumpanytype(?); END;"); cstmt.registerOutParameter(1,OracleTypes.VARCHAR); // obj is the instance of TypeDescriptor that you have retrieved cstmt.setObject(2,obj); cstmt.execute(); String str = (String)cstmt.getObject(1); ...

The oracle.sql.ANYDATA class enables you to access SYS.ANYDATA instances from the database. An instance of this class can be obtained from any valid instance of oracle.sql.Datum class. The convertDatum factory method takes an instance of Datum and returns an instance of ANYDATA. The syntax for this factory method is as follows:
public static ANYDATA convertDatum(Datum datum) throws SQLException

The following is sample code for creating an instance of oracle.sql.ANYDATA:

// struct is a valid instance of oracle.sql.STRUCT that either comes from the // database or has been constructed in Java. ANYDATA myAnyData = ANYDATA.convertDatum(struct);

Example 44 Accessing an Instance of ANYDATA from the Database ... // anydata_table has been created as: // CREATE TABLE anydata_tab (data SYS.ANYDATA) Statement stmt = conn.createStatement(); ResultSet rs = stmt.executeQuery("select data from my_anydata_tab"); while(rs.next()) { ANYDATA anydata = (ANYDATA)rs.getObject(1); if(!anydata.isNull()) { TypeDescriptor td = anydata.getTypeDescriptor(); if(td.getTypeCode() == OracleType.TYPECODE_OBJECT) STRUCT struct = (STRUCT)anydata.accessDatum(); } } ...

Example 45 Inserting an Object as ANYDATA in a Database Table

Consider the following table and object type definition:

CREATE TABLE anydata_tab ( id NUMBER, data SYS.ANYDATA) CREATE OR REPLACE TYPE employee AS OBJECT ( empno NUMBER, ename VARCHAR2(10) )

4-18 Oracle Database JDBC Developer's Guide and Reference

Additional Oracle Type Extensions

To create an instance of the EMPLOYEE SQL object type and to insert it into anydata_tab:
... PreparedStatement pstmt = conn.prepareStatement("insert into anydata_table values (?,?)"); StructDescriptor sd = StructDescriptor.createDescriptor("EMPLOYEE",(OracleConnection)conn); Object[] objattr = new Object[2]; objattr[0] = new BigDecimal(1120); objattr[1] = new String("Papageno"); STRUCT myEmployeeStr = new STRUCT(sd,conn,objattr); ANYDATA anyda = ANYDATA.convertDatum(myEmployeeStr); pstmt.setInt(1,123); pstmt.setObject(2,anyda); pstmt.executeUpdate(); ...

Example 46 Selecting an ANYDATA Column from a Database Table ... Statement stmt = conn.createStatement(); ResultSet rs = stmt.executeQuery("select data from anydata_table"); while(rs.next()) { ANYDATA obj = (ANYDATA)rs.getObject(1); TypeDescriptor td = obj.getTypeDescriptor(); } rs.close(); stmt.close(); ...

The oracle.jdbc Package

The interfaces of the oracle.jdbc package define the Oracle extensions to the interfaces in java.sql. These extensions provide access to SQL-format data as described in this chapter. They also provide access to other Oracle-specific functionality, including Oracle performance enhancements. For the oracle.jdbc package, Table 42 lists key interfaces and classes used for connections, statements, and result sets.
Table 42 Name OracleDriver OracleConnection Key Interfaces and Classes of the oracle.jdbc Package Interface or Class Class Interface Key Functionality Implements java.sql.Driver Provides methods to start and stop an Oracle Database instance and to return Oracle statement objects and methods to set Oracle performance extensions for any statement run in the current connection. Implements java.sql.Connection.

Oracle Extensions 4-19

Additional Oracle Type Extensions

Table 42 (Cont.) Key Interfaces and Classes of the oracle.jdbc Package Name OracleStatement Interface or Class Interface Key Functionality Provides methods to set Oracle performance extensions for individual statement. Is a supertype of OraclePreparedStatement and OracleCallableStatement. Implements java.sql.Statement. OraclePreparedStatement Interface Provides setXXX methods to bind oracle.sql.* types into a prepared statement. Implements java.sql.PreparedStatement. Extends OracleStatement. Is a supertype of OracleCallableStatement. OracleCallableStatement Interface Provides getXXX methods to retrieve data in oracle.sql format and setXXX methods to bind oracle.sql.* types into a callable statement. Implements java.sql.CallableStatement. Extends OraclePreparedStatement. OracleResultSet Interface Provides getXXX methods to retrieve data in oracle.sql format. Implements java.sql.ResultSet. OracleResultSetMetaData Interface Provides methods to get metadata information about Oracle result sets, such as column names and data types. Implements java.sql.ResultSetMetaData. OracleDatabaseMetaData Class Provides methods to get metadata information about the database, such as database product name and version, table information, and default transaction isolation level. Implements java.sql.DatabaseMetaData). OracleTypes Class Defines integer constants used to identify SQL types. For standard types, it uses the same values as the standard java.sql.Types class. In addition, it adds constants for Oracle extended types.

This section covers the following topics:

Interface oracle.jdbc.OracleConnection Interface oracle.jdbc.OracleStatement Interface oracle.jdbc.OraclePreparedStatement

4-20 Oracle Database JDBC Developer's Guide and Reference

Additional Oracle Type Extensions

Interface oracle.jdbc.OracleCallableStatement Interface oracle.jdbc.OracleResultSet Interface oracle.jdbc.OracleResultSetMetaData Class oracle.jdbc.OracleTypes Method getJavaSqlConnection

Interface oracle.jdbc.OracleConnection
This interface extends standard JDBC connection functionality to create and return Oracle statement objects, set flags and options for Oracle performance extensions, support type maps for Oracle objects, and support client identifiers. In Oracle Database 11g Release 1 (11.1), new methods have been added to this interface that enable the starting up and shutting down of an Oracle Database instance. Also, for better visibility and clarity, all connection properties are defined as constants in the OracleConnection interface. This interface also defines factory methods for constructing oracle.sql data values like DATE and NUMBER. Remember the following points while using factory methods:

All code that constructs instances of the oracle.sql types should use the Oracle extension factory methods. For example, ARRAY, BFILE, DATE, INTERVALDS, NUMBER, STRUCT, TIME, TIMESTAMP, and so on. All code that constructs instances of the standard types should use the JDBC 4.0 standard factory methods. For example, CLOB, BLOB, NCLOB, and so on. There are no factory methods for CHAR, JAVA_STRUCT, ArrayDescriptor, and StructDescriptor. These types are for internal driver use only.

Client Identifiers In a connection pooling environment, the client identifier can be used to identify the lightweight user using the database session currently. A client identifier can also be used to share the Globally Accessed Application Context between different database sessions. The client identifier set in a database session is audited when database auditing is turned on.
See Also:

Oracle Database Advanced Application Developer's Guide

The following oracle.jdbc.OracleConnection methods are Oracle-defined extensions:

cancel Performs an immediate (asynchronous) termination of any currently executing operation on this connection

commit Commits the transaction with the given options

getDefaultExecuteBatch Retrieves the default update-batching value for this connection

setDefaultExecuteBatch Sets the default update-batching value for this connection

getDefaultRowPrefetch

Oracle Extensions 4-21

Additional Oracle Type Extensions

Retrieves the default row-prefetch value for this connection

setDefaultRowPrefetch Sets the default row-prefetch value for this connection

Interface oracle.jdbc.OracleStatement
This interface extends standard JDBC statement functionality and is the superinterface of the OraclePreparedStatement and OracleCallableStatement classes. Extended functionality includes support for setting flags and options for Oracle performance extensions on a statement-by-statement basis, as opposed to the OracleConnection interface that sets these on a connectionwide basis. The following oracle.jdbc.OracleStatement methods are Oracle-defined extensions:

defineColumnType Defines the type you will use to retrieve data from a particular database table column
Note:

This method is no longer needed or recommended for use with the JDBC Thin driver.

getRowPrefetch Retrieves the row-prefetch value for this statement

setRowPrefetch Sets the row-prefetch value for this statement

Interface oracle.jdbc.OraclePreparedStatement
This interface extends the OracleStatement interface and extends standard JDBC prepared statement functionality. Also, the oracle.jdbc.OraclePreparedStatement interface is extended by the OracleCallableStatement interface. Extended functionality consists of setXXX methods for binding oracle.sql.* types and objects to prepared statements, and methods to support Oracle performance extensions on a statement-by-statement basis.
Note: Do not use the PreparedStatement interface to create a trigger that refers to a :NEW or :OLD column. Use Statement instead. Using PreparedStatement will cause execution to fail with the message java.sql.SQLException: Missing IN or OUT parameter at index:: 1

Interface oracle.jdbc.OracleCallableStatement
This interface extends the OraclePreparedStatement interface, which extends the OracleStatement interface and incorporates standard JDBC callable statement functionality.

4-22 Oracle Database JDBC Developer's Guide and Reference

Additional Oracle Type Extensions

Note: Do not use the CallableStatement interface to create a trigger that refers to a :NEW or :OLD column. Use Statement instead; using CallableStatement will cause execution to fail with the message java.sql.SQLException: Missing IN or OUT parameter at index::1

Note:

The setXXX(String,...) and registerOutParameter(String,...) methods can be used only if all binds are procedure or function parameters only. The statement can contain no other binds and the parameter binds must be indicated with a question mark (?) and not :XX. If you are using setXXX(int,...), setXXXAtName(String,...) or a combination of both, then any output parameter is bound with registerOutParameter(int,...) and not registerOutParameter(String,...), which is for named parameter notation.

Interface oracle.jdbc.OracleResultSet
This interface extends standard JDBC result set functionality, implementing getXXX methods for retrieving data into oracle.sql.* objects.

Interface oracle.jdbc.OracleResultSetMetaData
This interface extends standard JDBC result set metadata functionality to retrieve information about Oracle result set objects.
See Also:

"Using Result Set Metadata Extensions" on page 11-14

Class oracle.jdbc.OracleTypes
The OracleTypes class defines constants that JDBC uses to identify SQL types. Each variable in this class has a constant integer value. The oracle.jdbc.OracleTypes class duplicates the type code definitions of the standard Java java.sql.Types class and contains these additional type codes for Oracle extensions:

OracleTypes.BFILE OracleTypes.ROWID OracleTypes.CURSOR (for REF CURSOR types)

As in java.sql.Types, all the variable names are in uppercase text. JDBC uses the SQL types identified by the elements of the OracleTypes class in two main areas: registering output parameters and in the setNull method of the PreparedStatement class. OracleTypes and Registering Output Parameters The type codes in java.sql.Types or oracle.jdbc.OracleTypes identify the SQL types of the output parameters in the registerOutParameter method of the

Oracle Extensions 4-23

Additional Oracle Type Extensions

java.sql.CallableStatement and oracle.jdbc.OracleCallableStatement interfaces. These are the forms that the registerOutputParameter method can take for the CallableStatement and OracleCallableStatement interfaces
cs.registerOutParameter(int index, int sqlType); cs.registerOutParameter(int index, int sqlType, String sql_name); cs.registerOutParameter(int index, int sqlType, int scale);

In these signatures, index represents the parameter index, sqlType is the type code for the SQL data type, sql_name is the name given to the data type, for user-defined types, when sqlType is a STRUCT, REF, or ARRAY type code, and scale represents the number of digits to the right of the decimal point, when sqlType is a NUMERIC or DECIMAL type code. The following example uses a CallableStatement interface to call a procedure named charout, which returns a CHAR data type. Note the use of the OracleTypes.CHAR type code in the registerOutParameter method.
CallableStatement cs = conn.prepareCall ("BEGIN charout (?); END;"); cs.registerOutParameter (1, OracleTypes.CHAR); cs.execute (); System.out.println ("Out argument is: " + cs.getString (1));

The next example uses a CallableStatement interface to call structout, which returns a STRUCT data type. The form of registerOutParameter requires you to specify the type code, Types.STRUCT or OracleTypes.STRUCT, as well as the SQL name, EMPLOYEE. The example assumes that no type mapping has been declared for the EMPLOYEE type, so it is retrieved into a STRUCT data type. To retrieve the value of EMPLOYEE as an oracle.sql.STRUCT object, the statement object cs is cast to OracleCallableStatement and the Oracle extension getSTRUCT method is invoked.
CallableStatement cs = conn.prepareCall ("BEGIN structout (?); END;"); cs.registerOutParameter (1, OracleTypes.STRUCT, "EMPLOYEE"); cs.execute (); // get the value into a STRUCT because it // is assumed that no type map has been defined STRUCT emp = ((OracleCallableStatement)cs).getSTRUCT (1);

OracleTypes and the setNull Method The type codes in Types and OracleTypes identify the SQL type of the data item, which the setNull method sets to NULL. The setNull method can be found in the java.sql.PreparedStatement and oracle.jdbc.OraclePreparedStatement interfaces. These are the forms that the setNull method can take for the PreparedStatement and OraclePreparedStatement objects:
ps.setNull(int index, int sqlType); ps.setNull(int index, int sqlType, String sql_name);

In these signatures, index represents the parameter index, sqlType is the type code for the SQL data type, and sql_name is the name given to the data type, for
4-24 Oracle Database JDBC Developer's Guide and Reference

Additional Oracle Type Extensions

user-defined types, when sqlType is a STRUCT, REF, or ARRAY type code. If you enter an invalid sqlType, a ParameterTypeConflict exception is thrown. The following example uses a prepared statement to insert a null value into the database. Note the use of OracleTypes.NUMERIC to identify the numeric object set to NULL. Alternatively, Types.NUMERIC can be used.
PreparedStatement pstmt = conn.prepareStatement ("INSERT INTO num_table VALUES (?)"); pstmt.setNull (1, OracleTypes.NUMERIC); pstmt.execute ();

In this example, the prepared statement inserts a NULL STRUCT object of type EMPLOYEE into the database.
PreparedStatement pstmt = conn.prepareStatement ("INSERT INTO employee_table VALUES (?)"); pstmt.setNull (1, OracleTypes.STRUCT, "EMPLOYEE"); pstmt.execute ();

Method getJavaSqlConnection
The getJavaSqlConnection method of the oracle.sql.* classes returns java.sql.Connection. This method is available for the following Oracle data type classes:
Note: The getConnection method used in Oracle 8i and earlier versions of JDBC driver returns oracle.jdbc.driver.OracleConnection. The use of the classes in the oracle.jdbc.driver package was deprecated in favor of the oracle.jdbc package in Oracle 9i release. In Oracle Database 11g Release 1 (11.1), the classes in the package oracle.jdbc.driver have been desupported.

oracle.sql.ARRAY oracle.sql.BFILE oracle.sql.BLOB oracle.sql.CLOB oracle.sql.OPAQUE oracle.sql.REF oracle.sql.STRUCT

The following code snippet shows the getJavaSqlConnection method in the Array class:
public class ARRAY { java.sql.Connection getJavaSqlConnection() throws SQLException; ... }

Oracle Extensions 4-25

DML Returning

DML Returning
The DML returning feature provides more functionality compared to retrieval of auto-generated keys. It can be used to retrieve not only auto-generated keys, but also other columns or values that the application may use.
Note:

The server-side internal driver does not support DML returning and retrieval of auto-generated keys. You cannot use both DML returning and retrieval of auto-generated keys in the same statement.

The following sections explain the support for DML returning:

Oracle-Specific APIs Running DML Returning Statements Example of DML Returning Limitations of DML Returning
See Also:

"Retrieval of Auto-Generated Keys" on page 3-4

Oracle-Specific APIs
The OraclePreparedStatement interface is enhanced with Oracle-specific application programming interfaces (APIs) to support DML returning. The registerReturnParameter and getReturnResultSet methods have been added to the oracle.jdbc.OraclePreparedStatement interface, to register parameters that are returned and data retrieved by DML returning. The registerReturnParameter method is used to register the return parameter for DML returning. The method throws a SQLException instance if an error occurs. You must pass a positive integer specifying the index of the return parameter. You also must specify the type of the return parameter. You can also specify the maximum bytes or characters of the return parameter. This method can be used only with char or RAW types. You can also specify the fully qualified name of a SQL structure type.
Note:

If you do not know the maximum size of the return parameters, then you should use registerReturnParameter(int paramIndex, int externalType), which picks the default maximum size. If you know the maximum size of return parameters, using registerReturnParameter(int paramIndex, int externalType, int maxSize) can reduce memory consumption.

The getReturnResultSet method fetches the data returned from DML returning and returns it as a ResultSet object. The method throws a SQLException exception if an error occurs.
Note:

The Oracle-specific APIs for the DML returning feature are in ojdbc5.jar for Java Development Kit (JDK) 1.5 and in ojdbc6.jar for JDK 1.6.

4-26 Oracle Database JDBC Developer's Guide and Reference

DML Returning

Running DML Returning Statements

Before running a DML returning statement, the JDBC application must call one or more of the registerReturnParameter methods. The method provides the JDBC drivers with information, such as type and size, of the return parameters. The DML returning statement is then processed using one of the standard JDBC APIs, executeUpdate or execute. You can then fetch the returned parameters as a ResultSet object using the getReturnResultSet method of the oracle.jdbc.OraclePreparedStatement interface. In order to read the values in the ResultSet object, the underlying Statement object must be open. When the underlying Statement object is closed, the returned ResultSet object is also closed. This is consistent with ResultSet objects that are retrieved by processing SQL query statements. When a DML returning statement is run, the concurrency of the ResultSet object returned by the getReturnResultSet method must be CONCUR_READ_ONLY and the type of the ResultSet object must be TYPE_FORWARD_ONLY or TYPE_SCROLL_INSENSITIVE.

Example of DML Returning

This section provides two code examples of DML returning. The following code example illustrates the use of DML returning. In this example, assume that the maximum size of the name column is 100 characters. Because the maximum size of the name column is known, the registerReturnParameter(int paramIndex, int externalType, int maxSize) method is used.
... OraclePreparedStatement pstmt = (OraclePreparedStatement)conn.prepareStatement( "delete from tab1 where age < ? returning name into ?"); pstmt.setInt(1,18); /** register returned parameter * in this case the maximum size of name is 100 chars */ pstmt.registerReturnParameter(2, OracleTypes.VARCHAR, 100); // process the DML returning statement count = pstmt.executeUpdate(); if (count>0) { ResultSet rset = pstmt.getReturnResultSet(); //rest is not null and not empty while(rset.next()) { String name = rset.getString(1); ... } } ...

The following code example also illustrates the use of DML returning. However, in this case, the maximum size of the return parameters is not known. Therefore, the registerReturnParameter(int paramIndex, int externalType) method is used.
... OraclePreparedStatement pstmt = (OraclePreparedStatement)conn.prepareStatement( "insert into lobtab values (100, empty_clob()) returning col1, col2 into ?, ?");

Oracle Extensions 4-27

Accessing PL/SQL Index-by Tables

// register return parameters pstmt.registerReturnParameter(1, OracleTypes.INTEGER); pstmt.registerReturnParameter(2, OracleTypes.CLOB); // process the DML returning SQL statement pstmt.executeUpdate(); ResultSet rset = pstmt.getReturnResultSet(); int r; CLOB clob; if (rset.next()) { r = rset.getInt(1); System.out.println(r); clob = (CLOB)rset.getClob(2); ... } ...

Limitations of DML Returning

When using DML returning, be aware of the following:

It is unspecified what the getReturnResultSet method returns when it is invoked more than once. You should not rely on any specific action in this regard. The ResultSet objects returned from the execution of DML returning statements do not support the ResultSetMetaData type. Therefore, the applications must know the information of return parameters before running DML returning statements. Streams are not supported with DML returning. DML returning cannot be combined with batch update. You cannot use both the auto-generated key feature and the DML returning feature in a single SQL DML statement. For example, the following is not allowed:
... PreparedStatement pstmt = conn.prepareStatement(insert into orders (?, ?, ?) returning order_id into ?"); pstmt.setInt(1, seq01.NEXTVAL); pstmt.setInt(2, 100); pstmt.setInt(3, 966431502); pstmt.registerReturnParam(4, OracleTypes.INTEGER); pstmt.executeUpdate; ResultSet rset = pstmt.getGeneratedKeys; ...

Accessing PL/SQL Index-by Tables

Oracle JDBC drivers enable JDBC applications to make PL/SQL calls with index-by table parameters. This section covers the following topics:

Overview Binding IN Parameters Receiving OUT Parameters

4-28 Oracle Database JDBC Developer's Guide and Reference

Accessing PL/SQL Index-by Tables

Type Mappings
Note:

Index-by tables of PL/SQL records are not supported.

Overview
Oracle JDBC drivers support PL/SQL index-by tables of scalar data types. Table 43 displays the supported scalar types and the corresponding JDBC type codes.
Table 43 PL/SQL Types and Corresponding JDBC Types JDBC Types NUMERIC NUMERIC NUMERIC NUMERIC NUMERIC NUMERIC NUMERIC VARCHAR

PL/SQL Types BINARY_INTEGER NATURAL NATURALN PLS_INTEGER POSITIVE POSITIVEN SIGNTYPE STRING

Note: Oracle JDBC does not support RAW, DATE, and PL/SQL RECORD as element types.

Typical Oracle JDBC input binding, output registration, and data access methods do not support PL/SQL index-by tables. This chapter introduces additional methods to support these types. The OraclePreparedStatement and OracleCallableStatement classes define the additional methods. These methods include the following:

setPlsqlIndexTable registerIndexTableOutParameter getOraclePlsqlIndexTable getPlsqlIndexTable

These methods handle PL/SQL index-by tables as IN, OUT, or IN OUT parameters, including function return values.
See Also:

Oracle Database PL/SQL Language Reference

Binding IN Parameters
To bind a PL/SQL index-by table parameter in the IN parameter mode, use the setPlsqlIndexTable method defined in the OraclePreparedStatement and OracleCallableStatement classes.
synchronized public void setPlsqlIndexTable (int paramIndex, Object arrayData, int maxLen, int curLen, int elemSqlType, int elemMaxLen) throws SQLException

Oracle Extensions 4-29

Accessing PL/SQL Index-by Tables

Table 44 describes the arguments of the setPlsqlIndexTable method.

Table 44 Argument int paramIndex Object arrayData Arguments of the setPlsqlIndexTable Method Description Indicates the parameter position within the statement. Is an array of values to be bound to the PL/SQL index-by table parameter. The value is of type java.lang.Object. The value can be a Java primitive type array, such as int[], or a Java object array, such as BigDecimal[]. Specifies the maximum table length of the index-by table bind value that defines the maximum possible curLen for batch updates. For standalone binds, maxLen should use the same value as curLen. This argument is required. Specifies the actual size of the index-by table bind value in arrayData. If the curLen value is smaller than the size of arrayData, then only the curLen number of table elements is passed to the database. If the curLen value is larger than the size of arrayData, then the entire arrayData is sent to the database. Specifies the index-by table element type based on the values defined in the OracleTypes class. Specifies the index-by table element maximum length in case the element type is CHAR, VARCHAR, or RAW. This value is ignored for other types.

int maxLen

int curLen

int elemSqlType int elemMaxLen

The following code example uses the setPlsqlIndexTable method to bind an index-by table as an IN parameter:
// Prepare the statement OracleCallableStatement procin = (OracleCallableStatement) conn.prepareCall ("begin procin (?); end;"); // index-by table bind value int[] values = { 1, 2, 3 }; // maximum length of the index-by table bind value. This // value defines the maximum possible "currentLen" for batch // updates. For standalone binds, "maxLen" should be the // same as "currentLen". int maxLen = values.length; // actual size of the index-by table bind value int currentLen = values.length; // index-by table element type int elemSqlType = OracleTypes.NUMBER; // index-by table element length in case the element type // is CHAR, VARCHAR or RAW. This value is ignored for other // types. int elemMaxLen = 0; // set the value procin.setPlsqlIndexTable (1, values, maxLen, currentLen, elemSqlType, elemMaxLen);

4-30 Oracle Database JDBC Developer's Guide and Reference

Accessing PL/SQL Index-by Tables

// execute the call procin.execute ();

Receiving OUT Parameters

This section describes how to register a PL/SQL index-by table as an OUT parameter. In addition, it describes how to access the OUT bind values in various mapping styles.
Note:

The methods described in this section apply to function return values and the IN OUT parameter mode as well.

Registering the OUT Parameters To register a PL/SQL index-by table as an OUT parameter, use the registerIndexTableOutParameter method defined in the OracleCallableStatement class.
synchronized public void registerIndexTableOutParameter (int paramIndex, int maxLen, int elemSqlType, int elemMaxLen) throws SQLException

Table 45 describes the arguments of the registerIndexTableOutParameter method.

Table 45 Argument int paramIndex int maxLen int elemSqlType int elemMaxLen Arguments of the registerIndexTableOutParameter Method Description Indicates the parameter position within the statement. Specifies the maximum table length of the index-by table bind value to be returned. Specifies the index-by table element type based on the values defined in the OracleTypes class. Specifies the index-by table element maximum length in case the element type is CHAR, VARCHAR, or FIXED_CHAR. This value is ignored for other types.

The following code example uses the registerIndexTableOutParameter method to register an index-by table as an OUT parameter:
// maximum length of the index-by table value. This // value defines the maximum table size to be returned. int maxLen = 10; // index-by table element type int elemSqlType = OracleTypes.NUMBER; // index-by table element length in case the element type // is CHAR, VARCHAR or FIXED_CHAR. This value is ignored for other // types int elemMaxLen = 0; // register the return value funcnone.registerIndexTableOutParameter (1, maxLen, elemSqlType, elemMaxLen);

Oracle Extensions 4-31

Accessing PL/SQL Index-by Tables

Accessing the OUT Parameter Values To access the OUT bind value, the OracleCallableStatement class defines multiple methods that return the index-by table values in different mapping styles. There are three mapping choices available in JDBC drivers:
Mappings JDBC default mappings Oracle mappings Methods to Use getPlsqlIndexTable(int) getOraclePlsqlIndexTable(int)

Java primitive type mappings getPlsqlIndexTable(int, Class)

Type Mappings
This section covers the following topics:

JDBC Default Mappings Oracle Mappings Java Primitive Type Mappings

JDBC Default Mappings The getPlsqlIndexTable(int) method returns index-by table elements using the JDBC default mappings. The syntax for this method is the following:
public Object getPlsqlIndexTable (int paramIndex) throws SQLException

Table 46 describes the argument of the getPlsqlIndexTable method.

Table 46 Argument int paramIndex Argument of the getPlsqlIndexTable Method Description This argument indicates the parameter position within the statement.

The return value is a Java array. The elements of this array are of the default Java type corresponding to the SQL type of the elements. For example, for an index-by table with elements of NUMERIC type code, the element values are mapped to BigDecimal by Oracle JDBC driver, and the getPlsqlIndexTable method returns a BigDecimal[] array. For a JDBC application, you must cast the return value to BigDecimal[] to access the table element values. The following code example uses the getPlsqlIndexTable method to return index-by table elements with JDBC default mapping:
// access the value using JDBC default mapping BigDecimal[] values = (BigDecimal[]) procout.getPlsqlIndexTable (1); // print the elements for (int i=0; i<values.length; i++) System.out.println (values[i].intValue());

Oracle Mappings The getOraclePlsqlIndexTable method returns index-by table elements using Oracle mapping.
4-32 Oracle Database JDBC Developer's Guide and Reference

Accessing PL/SQL Index-by Tables

public Datum[] getOraclePlsqlIndexTable (int paramIndex) throws SQLException

Table 47 describes the argument of the getOraclePlsqlIndexTable method.

Table 47 Argument int paramIndex Argument of the getOraclePlsqlIndexTable Method Description Indicates the parameter position within the statement.

The return value is an oracle.sql.Datum array, and the elements in the array are of the default Datum type corresponding to the SQL type of the element. For example, the element values of an index-by table of numeric elements are mapped to the oracle.sql.NUMBER type in Oracle mapping, and the getOraclePlsqlIndexTable method returns an oracle.sql.Datum array that contains oracle.sql.NUMBER elements. The following code example uses the getOraclePlsqlIndexTable method to access the elements of a PL/SQL index-by table OUT parameter, using Oracle mapping:
// Prepare the statement OracleCallableStatement procout = (OracleCallableStatement) conn.prepareCall ("begin procout (?); end;"); ... // run the call procout.execute (); // access the value using Oracle JDBC mapping Datum[] outvalues = procout.getOraclePlsqlIndexTable (1); // print the elements for (int i=0; i<outvalues.length; i++) System.out.println (outvalues[i].intValue());

Java Primitive Type Mappings The getPlsqlIndexTable(int, Class) method returns index-by table elements in Java primitive types. The return value is a Java array. The syntax for this method is the following:
synchronized public Object getPlsqlIndexTable (int paramIndex, Class primitiveType) throws SQLException

Table 48 describes the arguments of the getPlsqlIndexTable method.

Table 48 Argument int paramIndex Arguments of the getPlsqlIndexTable Method Description Indicates the parameter position within the statement.

Oracle Extensions 4-33

Accessing PL/SQL Index-by Tables

Table 48 (Cont.) Arguments of the getPlsqlIndexTable Method Argument Class primitiveType Description Specifies a Java primitive type to which the index-by table elements are to be converted. For example, if you specify java.lang.Integer.TYPE, the return value is an int array. The following are the possible values of this parameter: java.lang.Integer.TYPE java.lang.Long.TYPE java.lang.Float.TYPE java.lang.Double.TYPE java.lang.Short.TYPE

The following code example uses the getPlsqlIndexTable method to access the elements of a PL/SQL index-by table of numbers. In the example, the second parameter specifies java.lang.Integer.TYPE and the return value of the getPlsqlIndexTable method is an int array.
OracleCallableStatement funcnone = (OracleCallableStatement) conn.prepareCall ("begin ? := funcnone; end;"); // maximum length of the index-by table value. This // value defines the maximum table size to be returned. int maxLen = 10; // index-by table element type int elemSqlType = OracleTypes.NUMBER; // index-by table element length in case the element type // is CHAR, VARCHAR or RAW. This value is ignored for other // types int elemMaxLen = 0; // register the return value funcnone.registerIndexTableOutParameter (1, maxLen, elemSqlType, elemMaxLen); // execute the call funcnone.execute (); // access the value as a Java primitive array. int[] values = (int[]) funcnone.getPlsqlIndexTable (1, java.lang.Integer.TYPE); // print the elements for (int i=0; i<values.length; i++) System.out.println (values[i]);

4-34 Oracle Database JDBC Developer's Guide and Reference

5
Features Specific to JDBC Thin
This chapter introduces the Java Database Connectivity (JDBC) Thin client and covers the features supported only by the JDBC Thin driver. It also provides basic information about working with Oracle JDBC applets. This following topics are covered in this chapter:

Overview of JDBC Thin Client Additional Features Supported JDBC in Applets

Overview of JDBC Thin Client

The JDBC Thin client is a pure Java, Type IV driver. It is lightweight and easy to install. It provides high performance, comparable to the performance provided by the JDBC Oracle Call Interface (OCI) driver. The JDBC Thin driver is written entirely in Java, and therefore, it is platform-independent. Also, this driver does not require any additional Oracle software on the client-side. The JDBC Thin driver communicates with the server using TTC, a protocol developed by Oracle to access data from Oracle Database. It can be used for application servers as well as for applets. The driver allows a direct connection to the database by providing an implementation of TCP/IP that implements Oracle Net and TTC on top of Java sockets. Both of these protocols are lightweight implementation versions of their counterparts on the server. The Oracle Net protocol runs over TCP/IP only. The JDBC Thin driver can be used on both the client-side and the server-side. On the client-side, drivers can be used in Java applications or Java applets that run either on the client or in the middle tier of a three-tier configuration. On the server-side, this driver is used to access a remote Oracle Database instance or another session on the same database.

Additional Features Supported

The JDBC Thin driver supports all standard JDBC features. The JDBC Thin driver also provides support for the following additional features:

Support for Applets Default Support for Native XA

Features Specific to JDBC Thin 5-1

JDBC in Applets

Support for Applets

The JDBC Thin driver is the only Oracle JDBC driver that provides support for applets. This driver can be downloaded along with the Java applet that is being run in a browser.
Note:

When the JDBC Thin driver is used with an applet, the browser used on the client-side must have the capability to support Java sockets.

The HTTP protocol, which is usually used for communication over a network, is stateless. However, the JDBC Thin driver is not stateless. Therefore, the initial HTTP request to download the applet and the JDBC Thin driver is stateless. After the JDBC Thin driver establishes the database connection, the communication between the browser and the database is stateful and in a two-tier configuration.
See Also:

"JDBC in Applets"

Default Support for Native XA

Similar to the JDBC OCI driver, the JDBC Thin driver also provides support for Native XA. However, the JDBC Thin driver provides support for Native XA by default. This is unlike the case of the JDBC OCI driver, in which the support for Native XA is not enabled by default.
See Also:

"Native-XA in Oracle JDBC Drivers" on page 28-20

JDBC in Applets
You can use only the Oracle JDBC Thin driver for an applet. This section describes what you must do to connect an applet to a database. This description includes how to use the Connection Manager feature of Oracle Database, or signed applets if you are connecting to a database that is running on a different host from the Web server. It also describes how your applet can connect to a database through a firewall. The section concludes with how to package and deploy the applet. The following topics are covered:

Connecting to the Database Through the Applet Connecting to a Database on a Different Host Than the Web Server Using Applets with Firewalls Packaging Applets Specifying an Applet in an HTML Page

Connecting to the Database Through the Applet

The most common task of an applet using the JDBC driver is to connect to and query a database. Because of applet security restrictions, unless particular steps are taken, an applet can open TCP/IP sockets only to the host from which it was downloaded. This is the host on which the Web server is running. This means that without these steps, your applet can connect only to a database that is running on the same host as the Web server.

5-2 Oracle Database JDBC Developer's Guide and Reference

JDBC in Applets

If your database and Web server are running on the same host, then there is no issue and no special steps are required. You can connect to the database as you would from an application. As with connecting from an application, there are two ways in which you can specify the connection information to the driver. You can provide it in the form of host:port:sid or in the form of TNS keyword-value syntax. For example, if the database to which you want to connect resides on host prodHost, at port 1521, and system identifier (SID) ORCL, and you want to connect with user name scott and password tiger, then use either of the two following connection strings:

Using host:port:sid syntax:

String connString="jdbc:oracle:thin:@prodHost:1521:ORCL"; OracleDataSource ods = new OracleDataSource(); ods.setURL(connString); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection();

Using TNS keyword-value syntax:

String connString = "jdbc:oracle:thin:@(description=(address_list= (address=(protocol=tcp)(port=1521)(host=prodHost))) OracleDataSource ods = new OracleDataSource(); ods.setURL(connString); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection(); (connect_data=(INSTANCE_NAME=ORCL)))";

If you use the TNS keyword/value pair to specify the connection information to the JDBC Thin driver, then you must declare the protocol as TCP. However, a Web server and database server both require many resources. You seldom find both servers running on the same computer. Usually, your applet connects to a database on a host other than the one on which the Web server runs. If you want your applet to connect to a database running on a different computer, then you have the following options:

Use the Oracle Connection Manager on the host computer. The applet can connect to the Connection Manager, which connects to a database on another computer. Use signed applets, which can request socket connection privileges to other computers.

Your applet can also take advantage of the data encryption and integrity checksum features of the Advanced Security option of Oracle Database.

Connecting to a Database on a Different Host Than the Web Server

If you are connecting to a database on a host other than the one on which the Web server is running, then you must overcome applet security restrictions. You can do this in the following ways:

Using the Oracle Connection Manager

Features Specific to JDBC Thin 5-3

JDBC in Applets

Using Signed Applets

Using the Oracle Connection Manager

The Oracle Connection Manager is a lightweight, highly scalable program that can receive Oracle Net packets and retransmit them to a different server. To a client running Oracle Net, the Connection Manager looks exactly like a database server. An applet that uses the JDBC Thin driver can connect to a Connection Manager running on the Web server host and have the Connection Manager redirect the Oracle Net packets to an Oracle server running on a different host. Figure 51 illustrates the relationship between the applet, the Oracle Connection Manager, and the database.
Figure 51 Applet, Connection Manager, and Database Relationship

Applet in Browser

TCP/IP (only)

CMAN

any Oracle Net protocol

Oracle Net Listener

Web Server

webHost

oraHost

Using the Oracle Connection Manager requires two steps:

Install and run the Connection Manager. Write the connection string that targets the Connection Manager.

Installing and Running the Oracle Connection Manager You must install the Connection Manager, available on the Oracle distribution media, onto the Web server host. On the Web server host, create a CMAN.ORA file in the ORACLE_HOME/NET8/ADMIN directory. The options you can declare in a CMAN.ORA file include firewall and connection pooling support. Here is an example of a very simple CMAN.ORA file. Replace web-server-host with the name of your Web server host. The fourth line in the file indicates that the Connection Manager is listening on port 1610. You must use the same port number in your connection string for JDBC.
cman = (ADDRESS_LIST = (ADDRESS = (PROTOCOL=TCP) (HOST=web-server-host) (PORT=1610))) cman_profile = (parameter_list = (MAXIMUM_RELAYS=512) (LOG_LEVEL=1) (TRACING=YES) (RELAY_STATISTICS=YES) (SHOW_TNS_INFO=YES) (USE_ASYNC_CALL=YES) (AUTHENTICATION_LEVEL=0) )

After you create the file, start the Connection Manager at the operating system prompt with the following command:
5-4 Oracle Database JDBC Developer's Guide and Reference

JDBC in Applets

cmctl start

To use your applet, you must now write the connection string for it. Writing the URL that Targets the Connection Manager The following text describes how to write the URL in your applet, so that the applet connects to the Connection Manager and the Connection Manager connects with the database. In the URL, you specify an address list that lists the protocol, port, and name of the Web server host on which the Connection Manager is running, followed by the protocol, port, and name of the host on which the database is running. The following example describes the configuration illustrated in Figure 51. The Web server on which the Connection Manager is running is on host webHost and is listening on port 1610. The database to which you want to connect is running on host oraHost, listening on port 1521, and SID ORCL. You write the URL in TNS keyword-value format:
String myURL = "jdbc:oracle:thin:@(description=(address_list= (address=(protocol=tcp)(port=1610)(host=webHost)) (address=(protocol=tcp)(port=1521)(host=oraHost))) (connect_data=(INSTANCE_NAME=orcl)) (source_route=yes))"; OracleDataSource ods = new OracleDataSource(); ods.setURL(myURL); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection();

The first element in the address_list entry represents the connection to the Connection Manager. The second element represents the database to which you want to connect. The order in which you list the addresses is important. When your applet uses a URL, such as the preceding one, it will function exactly as if it were connected directly to the database on the host oraHost. Connecting Through Multiple Connection Managers Your applet can reach its target database even if it first has to go through multiple Connection Managers. For example, if the Connection Managers form a proxy chain. To do this, add the addresses of the Connection Managers to the address list, in the order that you plan to access them. The database listener should be the last address on this list.

Using Signed Applets

In a Java Development Kit (JDK) 1.2.x-based or later browser, an applet can request socket connection privileges and connect to a database running on a different host than the Web server host. Starting from Netscape 4.0, you perform this by signing your applet, that is, writing a signed applet. You must follow these steps:
1.

Sign the applet. For information about the steps you must follow to sign an applet, refer to the Signed Applet Example at http://java.sun.com/developer/technicalArticles/Security/Sign ed/index.html

2.

Include applet code that asks for appropriate privileges before opening a socket. If you are using Netscape, then your code would include a statement like this:

Features Specific to JDBC Thin 5-5

JDBC in Applets

netscape.security.PrivilegeManager.enablePrivilege("UniversalConnect"); OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:oracle:thin:scott/tiger@dlsun511:1721:orcl"); Connection conn = ods.getConnection(); 3.

You must obtain an object-signing certificate. Refer to a site that provides information about obtaining and installing a certificate.

For information about the Java Security API, including signed applet examples, see the following Sun Microsystems site: http://java.sun.com/security

Using Applets with Firewalls

Under standard circumstances, an applet that uses the JDBC Thin driver cannot access the database through a firewall. In general, the purpose of a firewall is to prevent unauthorized clients from reaching the server. In the case of applets trying to connect to the database, the firewall prevents the opening of a TCP/IP socket to the database. In general, firewalls are rule-based. They have a list of rules that define which clients can connect, and which cannot. Firewalls compare the host name of the client with the rules and, based on this comparison, either grant the client access or deny access. If the host name lookup fails, then the firewall tries again. This time, the firewall extracts the IP address of the client and compares it to the rules. The firewall is designed to do this so that users can specify rules that include host names as well as IP addresses. You can solve the firewall issue by using an Oracle Net-compliant firewall and connection strings that comply with the firewall configuration. Oracle Net-compliant firewalls are available from many leading vendors. An unsigned applet can access only the same host from which it is downloaded. In this case, the Oracle Net-compliant firewall must be installed on that host. In contrast, a signed applet can connect to any host. In this case, the firewall on the target host controls the access. Connecting through a firewall requires two steps, as described in the following sections:

Configuring a Firewall for Applets that use the JDBC Thin Driver Writing a URL to Connect Through a Firewall

Configuring a Firewall for Applets that use the JDBC Thin Driver
The instructions in this section assume that you are running an Oracle Net-compliant firewall. Java applets do not have access to the local system. Because of the security limitations, applets cannot access the host name or environment variables on the local system. As a result, the JDBC Thin driver cannot access the host name on which it is running. The firewall cannot be provided with the host name. To allow requests from JDBC Thin clients to go through the firewall, you must do the following to the list of firewall rules:

Add the IP address, and not the host name, of the host on which the JDBC applet is running. Ensure that the host name, "__jdbc__", never appears in the firewall rules. This host name has been hard-coded as a false host name inside the driver to force an IP address lookup. If you do enter this host name in the list of rules, then every applet using the JDBC Thin driver will be able to go through your firewall.

5-6 Oracle Database JDBC Developer's Guide and Reference

JDBC in Applets

Writing a URL to Connect Through a Firewall

To write a URL that enables you to connect through a firewall, you must specify the name of the firewall host and the name of the database host to which you want to connect. For example, if you want to connect to a database on host oraHost, listening on port 1521, with SID ORCL, and you are going though a firewall on host fireWallHost, listening on port 1610, then use the following URL:
OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:oracle:thin:" + "@(description=(address_list=" + (address=(protocol=tcp)(host=<firewall-host>)(port=1610))" + "(address=(protocol=tcp)(host=oraHost)(port=1521)))" + "(source_route=yes)" + "(connect_data=(SERVICE_NAME=orcl)))"); ); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection();

Note:

To connect through a firewall, you cannot specify the URL in host:port:sid syntax. For example, a URL specified as follows will not work:

String connString = "jdbc:oracle:thin:@example.us.oracle.com:1521:orcl"; OracleDataSource ods = new OracleDataSource(); ods.setURL(connString); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection();

The first element in the address_list represents the connection to the firewall. The second element represents the database to which you want to connect. Note that the order in which you specify the addresses is important. You can also write the preceding URL in the following format:
String connString = "jdbc:oracle:thin:@(description=(address_list= (address=(protocol=tcp)(port=1600)(host=fireWallHost)) (address=(protocol=tcp)(port=1521)(host=oraHost))) (connect_data=(INSTANCE_NAME=orcl)) (source_route=yes))"; OracleDataSource ods = new OracleDataSource(); ods.setURL(connString); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection();

When your applet uses a URL similar to the preceding URL, it will act as if it were connected to the database on host oraHost.

Features Specific to JDBC Thin 5-7

JDBC in Applets

Note:

All the parameters shown in the preceding example are required. In address_list, the firewall address must precede the database server address.

Packaging Applets
After you have coded your applet, you must package it and make it available to users. To package an applet, you will need your applet class files and the JDBC driver class files contained in the ojdbc5.jar or ojdbc6.jar files. Follow these steps:
1.

Move the JDBC driver classes file ojdbc5.jar or ojdbc6.jar to an empty directory. If your applet connects to a database with a non-US7ASCII and non-WE8ISO8859P1 character set and uses Oracle object types, then also move the orai18n.jar file to the same directory.

2. 3.

Add your applet classes files to the directory and any other files that the applet may require. Zip the applet classes and driver classes together into a single ZIP or Java Archive (JAR) file. The single ZIP file should contain the following:

Class files from the ojdbc5.jar or ojdbc6.jar files and required class files from the orai18n.jar files, if the applet requires Globalization Support Your applet classes

4.

Ensure that the ZIP or JAR file is not compressed.

You can now make the applet available to users. One way to do this is to add the APPLET tag to the HTML page from which the applet will be run. For example:
<APPLET WIDTH=500 HEIGHT=200 CODE=JdbcApplet ARCHIVE=JdbcApplet.zip CODEBASE=Applet_Samples </APPLET>

Specifying an Applet in an HTML Page

The APPLET tag specifies an applet that runs in the context of an HTML page. The APPLET tag can have the following attributes: CODE, ARCHIVE, CODEBASE, WIDTH, and HEIGHT. These attributes are described in the following sections:

CODE, HEIGHT, and WIDTH CODEBASE ARCHIVE

CODE, HEIGHT, and WIDTH

The HTML page that runs the applet must have an APPLET tag with an initial width and height to specify the size of the applet display area. You use the HEIGHT and WIDTH attributes to specify the size, measured in pixels. This size should not count any windows or dialog boxes that the applet opens. The APPLET tag must also specify the name of the file that contains the compiled applet. Specify the file name with the CODE attribute. Any path specified must be relative to the base URL of the applet. The path cannot be absolute.
5-8 Oracle Database JDBC Developer's Guide and Reference

JDBC in Applets

In the following example, JdbcApplet.class is the name of the compiled applet:

<APPLET CODE="JdbcApplet" WIDTH=500 HEIGHT=200> </APPLET>

If you use this form of the CODE attribute, then the classes for the applet and the JDBC Thin driver must be in the same directory as the HTML page.
Note: Do not include the file name extension, .class, in the CODE attribute.

CODEBASE
The CODEBASE attribute is optional. It specifies the base URL of the applet, that is, the name of the directory that contains the code of the applet. If it is not specified, then the URL of the document is used. This means that the classes for the applet and the JDBC Thin driver must be in the same directory as the HTML page. For example, if the current directory is my_Dir:
<APPLET WIDTH=500 HEIGHT=200 CODE=JdbcApplet CODEBASE="." </APPLET>

The attribute, CODEBASE=".", indicates that the applet resides in the current directory, my_Dir. Now, consider that the value of CODEBASE is set to Applet_Samples, as follows:
<APPLET WIDTH=500 HEIGHT=200 CODE=JdbcApplet CODEBASE="Applet_Samples" </APPLET>

This would indicate that the applet resides in the my_Dir/Applet_Samples directory.

ARCHIVE
The ARCHIVE attribute is optional. It specifies the name of the archive file that contains the applet classes and resources the applet needs. Oracle recommends using an archive file, which saves many extra round-trips to the server. The archive file will be preloaded. If you have more than one archive file in the list, separate them with commas. In the following example, the class files are stored in the archive file, JdbcApplet.zip:
<APPLET CODE="JdbcApplet" ARCHIVE="JdbcApplet.zip" WIDTH=500 HEIGHT=200> </APPLET>

Note:

Version 3.0 browsers do not support the ARCHIVE attribute.

Features Specific to JDBC Thin 5-9

JDBC in Applets

5-10 Oracle Database JDBC Developer's Guide and Reference

6
Features Specific to JDBC OCI Driver
This chapter introduces the features specific to the Java Database Connectivity (JDBC) Oracle Call Interface (OCI) driver. It also describes the OCI Instant Client. This chapter contains the following sections:

OCI Connection Pooling Transparent Application Failover OCI Native XA OCI Instant Client Instant Client Light (English)

OCI Connection Pooling

The OCI connection pooling feature is an Oracle-designed extension. The connection pooling provided by the JDBC OCI driver enables applications to have multiple logical connections, all of which are using a small set of physical connections. Each call on a logical connection is routed on to the physical connection that is available at the given time.
See Also:

Chapter 24, "OCI Connection Pooling"

Transparent Application Failover

The Transparent Application Failover feature of the JDBC OCI driver enables you to automatically reconnect to a database if the database instance to which the connection is made goes down. The new database connection, though created by a different node, is identical to the original.
See Also:

Chapter 27, "Transparent Application Failover"

OCI Native XA
The JDBC OCI driver also provides a feature called Native XA.
See Also:

"OCI Native XA" on page 28-20

OCI Instant Client

This section covers the following topics:

Overview of Instant Client

Features Specific to JDBC OCI Driver

6-1

OCI Instant Client

Benefits of Instant Client JDBC OCI Instant Client Installation Process Usage of Instant Client Patching Instant Client Shared Libraries Regeneration of Data Shared Library and ZIP files Database Connection Names for OCI Instant Client Environment Variables for OCI Instant Client

Overview of Instant Client

The Instant Client feature makes it extremely easy to deploy OCI, Oracle C++ Call Interface (OCCI), Open Database Connectivity (ODBC), and JDBC-OCI based customer applications, by eliminating the need for an Oracle home. The storage space requirement of a JDBC OCI application running in the Instant Client mode is significantly reduced compared to the same application running on a full client-side installation. The Instant Client shared libraries occupy only about one-fourth the disk space used by a full client installation. Table 61 shows the Oracle client-side files required to deploy a JDBC OCI application. Library names of release 11.1 are used in the table. The number part of library names will change in future releases to agree with the release.
Table 61 OCI Instant Client Shared Libraries
Description for Linux and UNIX Systems Microsoft Windows oci.dll oraociei11.dll orannzsbb11.dll ocijdbc11.dll All JDBC JAR files Description for Microsoft Windows Forwarding functions that applications link with Data and code Security Library OCI Instant Client JDBC Library See Also: "Check the Environment Variables" on page 2-3

Linux and UNIX Systems

libclnstsh.so.11.1 Client Code Library libociei.so libnnz11.so libocijdbc11.so ALL JDBC Java Archive (JAR) files OCI Instant Client Data Shared Library Security Library OCI Instant Client JDBC Library See Also: "Check the Environment Variables" on page 2-3

To provide Native XA functionality, you must copy the JDBC XA class library. On UNIX systems, this library, libheteroxa11.so, is located in the ORACLE_HOME/jdbc/lib directory. On Microsoft Windows, this library, heteroxa11.dll, is located in the ORACLE_HOME\bin directory.
Note:

Benefits of Instant Client

The benefits of Instant Client are the following:

Installation involves copying a small number of files. The number of required files and the total disk storage on the Oracle client-side are significantly reduced. There is no loss of functionality or performance for applications deployed with the Instant Client. It is simple for independent software vendors to package applications.

6-2 Oracle Database JDBC Developer's Guide and Reference

OCI Instant Client

JDBC OCI Instant Client Installation Process

The Instant Client libraries can be installed by choosing the Instant Client option from Oracle Universal Installer. The Instant Client libraries can also be downloaded from the Oracle Technology Network Web site. The installation process is as follows:
1. 2.

Download and install the Instant Client shared libraries and Oracle JDBC class libraries to a directory, such as instantclient. Set the library path environment variable to the directory from Step 1. For example, on UNIX systems, set the LD_LIBRARY_PATH environment variable to instantclient. On Microsoft Windows, set the PATH environment variable to locate the instantclient directory. Add the full path names of the JDBC class libraries to the CLASSPATH environment variable.

3.

After completing these steps you are ready to run the JDBC OCI application. The JDBC OCI application operates in the Instant Client mode when the OCI and JDBC shared libraries are accessible through the library path environment variable. In the Instant Client mode, there is no dependency on the ORACLE_HOME and none of the other code and data files provided in ORACLE_HOME is needed by JDBC OCI, except for the tnsnames.ora file. Instant Client can be also installed from Oracle Universal Installer by selecting the Instant Client option. The Instant Client files should always be installed in an empty directory. As with the OTN installation, you must set the LD_LIBRARY_PATH environment variable to the Instant Client directory to operate in the Instant Client mode. If you have done a complete client installation by choosing the Admin option, then the Instant Client shared libraries are also installed. The location of the Instant Client shared libraries and JDBC class libraries in a full client installation is: On Linux or UNIX systems:

libociei.so library is in $ORACLE_HOME/instantclient libclnstsh.so.11.1, libocijdbc11.so, and libnnz11.so are in $ORACLE_HOME/lib The JDBC class libraries are in $ORACLE_HOME/jdbc/lib

On Microsoft Windows:

oraociei11.dll library is in ORACLE_HOME\instantclient oci.dll, ocijdbc11.dll, and orannzsbb11.dll are in ORACLE_HOME\bin The JDBC class libraries are in ORACLE_HOME\jdbc\lib

By copying these files to a different directory, setting the library path to locate this directory, and adding the path names of the JDBC class libraries to the CLASSPATH environment variable, you can enable running the JDBC OCI application in the Instant Client mode.

Features Specific to JDBC OCI Driver

6-3

OCI Instant Client

Note:

To provide Native XA functionality, you must copy the JDBC XA class library. On UNIX, this library, libheteroxa11.so, is located in ORACLE_HOME/jdbc/lib. On Windows, this library, heteroxa11.dll, is located in ORACLE_HOME\bin. All the libraries must be copied from the same ORACLE_HOME and must be placed in the same directory. On hybrid platforms, such as Sparc64, if the JDBC OCI driver needs to be operated in the Instant Client mode, then you must copy the libociei.so library from the ORACLE_HOME/instantclient32 directory. You must copy all other Sparc64 libraries needed for the JDBC OCI Instant Client from the ORACLE_HOME/lib32 directory. Only one set of Oracle libraries should be specified in the library path environment variable. That is, if you have multiple directories containing Instant Client libraries, then only one such directory should be specified in the library path environment variable. If you have an Oracle home on your computer, then you should not have the ORACLE_HOME/lib and Instant Client directories in the library path environment variable simultaneously, regardless of the order in which they appear in the variable. That is, only one of ORACLE_HOME/lib directory (for non-Instant Client operation) or Instant Client directory (for Instant Client operation) should be specified in the library path environment variable. Oracle recommends that you download Instant Client from Oracle Technology Network (OTN)
http://www.oracle.com/technology/tech/oci/instantclien t/instantclient.html

Usage of Instant Client

Instant Client is a deployment feature and should be used for running production applications. For development, a full installation is necessary to access demonstration programs and so on. In general, all JDBC OCI functionality is available to an application being run in the Instant Client mode, except that the Instant Client mode is for client-side operation only. Therefore, server-side external procedures cannot operate in the Instant Client mode.

Patching Instant Client Shared Libraries

Because Instant Client is a deployment feature, the emphasis has been on reducing the number and size of files required to run a JDBC OCI application. Therefore, all files needed to patch Instant Client shared libraries are not available in an Instant Client deployment. An ORACLE_HOME based full client installation is needed to patch the Instant Client shared libraries. The opatch utility will take care of patching the Instant Client shared libraries.
Note:

On Microsoft Windows, you cannot patch the shared libraries.

6-4 Oracle Database JDBC Developer's Guide and Reference

OCI Instant Client

After applying the patch in an ORACLE_HOME environment, copy the files listed in Table 61, " OCI Instant Client Shared Libraries" to the instant client directory as described in "JDBC OCI Instant Client Installation Process". Instead of copying individual files, you can generate Instant Client ZIP files for OCI, OCCI, JDBC, and SQL*Plus as described in "Regeneration of Data Shared Library and ZIP files". Then, you can copy the ZIP files to the target computer and unzip them as described in "JDBC OCI Instant Client Installation Process". The opatch utility stores the patching information of the ORACLE_HOME installation in libclnstsh.so.11.1. This information can be retrieved by the following command:
genezi -v

Note that if the computer from where Instant Client is deployed does not have the genezi utility, then it must be copied from the ORACLE_HOME/bin directory on the computer that has the ORACLE_HOME installation.

Regeneration of Data Shared Library and ZIP files

The OCI Instant Client Data Shared Library, libociei.so, can be regenerated by performing the following steps in an Administrator Installation of ORACLE_HOME:
mkdir -p $ORACLE_HOME/rdbms/install/instantclient/light cd $ORACLE_HOME/rdbms/lib make -f ins_rdbms.mk ilibociei

A new version of the libociei.so Data Shared Library based on the current files in the ORACLE_HOME is then placed in the ORACLE_HOME/rdbms/install/instantclient directory. Note that the location of the regenerated Data Shared Library, libociei.so, is different from that of the original Data Shared Library, libociei.so, which is located in the ORACLE_HOME/instantclient directory. The preceding steps also generate Instant Client ZIP files for OCI, OCCI, JDBC, and SQL*Plus. Regeneration of data shared library and ZIP files is not available on Microsoft Windows platforms.

Database Connection Names for OCI Instant Client

All Oracle Net naming methods that do not require the ORACLE_HOME or TNS_ADMIN environment variables to locate configuration files, such as tnsnames.ora or sqlnet.ora, work in the Instant Client mode. In particular, the connection string can be specified in the following formats:

A Thin-style connection string of the form:

host:port:service_name

For example:
url="jdbc:oracle:oci:@//example.com:5521:bjava21"

A SQL connection URL string of the form:

//host:[port][/service name]

For example:
Features Specific to JDBC OCI Driver 6-5

OCI Instant Client

url="jdbc:oracle:oci:@//example.com:5521/bjava21

As an Oracle Net keyword-value pair. For example:

url="jdbc:oracle:oci:@(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp) (HOST=dlsun242) (PORT=5521)) (CONNECT_DATA=(SERVICE_NAME=bjava21)))"

Naming methods that require TNS_ADMIN to locate configuration files continue to work if the TNS_ADMIN environment variable is set.
See Also: Oracle Database Net Services Administrator's Guide for more information about connection formats

If the TNS_ADMIN environment variable is not set and TNSNAMES entries, such as inst1, are used, then the ORACLE_HOME environment variable must be set and the configuration files are expected to be in the $ORACLE_HOME/network/admin directory.
Note:

In this case, the ORACLE_HOME environment variable is used only for locating Oracle Net configuration files. No other component of Client Code Library uses the value of the ORACLE_HOME environment variable.

The empty connection string is not supported. However, an alternate way to use the empty connection string is to set the TWO_TASK environment variable on UNIX systems, or the LOCAL variable on Microsoft Windows, to either a tnsnames.ora entry or an Oracle Net keyword-value pair. If TWO_TASK or LOCAL is set to a tnsnames.ora entry, then the tnsnames.ora file must be loaded by the TNS_ADMIN or ORACLE_HOME setting. Example Consider that the listener.ora file on the database server contains the following information:
LISTENER = (ADDRESS_LIST=(ADDRESS=(PROTOCOL=tcp)(HOST=server6)(PORT=1573))) SID_LIST_LISTENER = (SID_LIST= (SID_DESC=(SID_NAME=rdbms3) (GLOBAL_DBNAME=rdbms3.server6.us.alchemy.com) (ORACLE_HOME=/home/dba/rdbms3/oracle)))

You can connect to this server in one of the following ways:

url = "jdbc:oracle:oci:@(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp) (HOST=server6)(PORT=1573)) (CONNECT_DATA=(SERVICE_NAME=rdbms3.server6.us.alchemy.com)))"

or:
url = "jdbc:oracle:oci:@//server6:1573/rdbms3.server6.us.alchemy.com"

Alternatively, you can set the TWO_TASK environment variable to any of the connection strings and connect to the database server without specifying the connection string along with the sqlplus command. For example, set the TWO_TASK environment in one of the following ways:
setenv TWO_TASK "(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=server6)(PORT=1573))

6-6 Oracle Database JDBC Developer's Guide and Reference

OCI Instant Client

(CONNECT_DATA=(SERVICE_NAME=rdbms3.server6.us.alchemy.com)))"

or:
setenv TWO_TASK //server6:1573/rdbms3.server6.us.alchemy.com

Now, you can connect to the database server using the following URL:
url = "jdbc:oracle:oci:@"

The connection string can also be stored in the tnsnames.ora file. For example, consider that the tnsnames.ora file contains the following:
conn_str = (DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=server6)(PORT=1573)) (CONNECT_DATA=(SERVICE_NAME=rdbms3.server6.us.alchemy.com)))

If this tnsnames.ora file is located in the /home/webuser/instantclient directory, then you can set the TNS_ADMIN environment variable (or LOCAL on Microsoft Windows) as follows:
setenv TNS_ADMIN /home/webuser/instantclient

Now, you can connect as follows:

url = "jdbc:oracle:oci:@conn_str"

Note:

The TNS_ADMIN environment variable specifies the directory where the tnsnames.ora file is located. However, TNS_ADMIN does not specify the full path of the tnsnames.ora file, instead it specifies the directory.

If this tnsnames.ora file is located in the /network/server6/home/dba/oracle/network/admin directory in the Oracle home, then instead of using TNS_ADMIN to locate the tnsnames.ora file, you can set the ORACLE_HOME environment variable as follows:
setenv ORACLE_HOME /network/server6/home/dba/oracle

Now, you can connect with either of the conn_str connection strings, as specified previously. If tnsnames.ora can be located by TNS_ADMIN or ORACLE_HOME, then TWO_TASK can be set to:
setenv TWO_TASK conn_str

You can then connect with the following URL:

url = "jdbc:oracle:oci:@"

Environment Variables for OCI Instant Client

The ORACLE_HOME environment variable no longer determines the location of the Globalization Support files and error message files. An OCI-only application does not require the ORACLE_HOME environment variable to be set. However, if the variable is set, then it does not have an impact on the operation of the OCI driver. OCI will always obtain its data from the Data Shared Library. If the Data Shared Library is not available, only then is the ORACLE_HOME environment variable used and a full client

Features Specific to JDBC OCI Driver

6-7

Instant Client Light (English)

installation is assumed. Even though the ORACLE_HOME environment variable is not required to be set, if it is set, then it must be set to a valid operating system path name that identifies a directory. Environment variables ORA_NLS10 and ORA_NLSPROFILES33 are ignored in the Instant Client mode. In the Instant Client mode, if the ORA_TZFILE variable is not set, then the smaller, default, timezone.dat file from the Data Shared Library is used. If the larger timezlrg.dat file is to be used from the Data Shared Library, then set the ORA_TZFILE environment variable to the name of the file without any absolute or relative path names. That is: On UNIX systems:
setenv ORA_TZFILE timezlrg.dat

On Microsoft Windows:
set ORA_TZFILE timezlrg.dat

If the driver is not operating in the Instant Client mode, then the ORA_TZFILE variable, if set, names a complete path name, as it does in previous Oracle Database releases. If TNSNAMES entries are used, then, as mentioned earlier, the TNS_ADMIN directory must contain the TNSNAMES configuration files, and if TNS_ADMIN is not set, then the ORACLE_HOME/network/admin directory must contain Oracle Net Services configuration files.

Instant Client Light (English)

The lightweight version of Instant Client is called Instant Client Light (English). Instant Client Light is the short name. Instant Client Light is a significantly smaller version of Instant Client. This reduces the disk space requirements of the client installation by about 63 MB. This is achieved by the lightweight Data Shared Library, libociicus.so on UNIX systems, which is 4 MB in size and a subset of the data shared library, libociei.so, which is 67 MB in size. The lightweight data shared library supports only a few character sets and error messages that are only in English. Therefore, the name Instant Client Light (English). Instant Client Light is designed for applications that require English-only error messages and use either US7ASCII, WE8DEC, or one of the Unicode character sets. Table 62 lists the names of the data shared libraries for Instant Client and Instant Client Light (English) on different platforms. The table also specifies the size of each data shared library in parentheses following the library file name.
Table 62 Platform Sun Solaris Linux Microsoft Windows Data Shared Library for Instant Client and Instant Client Light (English) Instant Client libociei.so (67 MB) libociei.so (67 MB) oraociei11.dll (85 MB) Instant Client Light (English) libociicus.so (4 MB) libociicus.so (4 MB) oraociicus10.dll (15 MB)

This section covers the following topics:

Globalization Settings

6-8 Oracle Database JDBC Developer's Guide and Reference

Instant Client Light (English)

Operation Installation

Globalization Settings
The NLS_LANG setting determines the language, territory, and character set as language_territory.characterset. In Instant Client Light, language can only be American, territory can be any that is supported, and characterset can be any one of the following:

Single-byte US7ASCII WE8DEC WE8MSWIN1252 WE8ISO8859P1

Unicode UTF8 AL16UTF16 AL32UTF8

Specifying character set or national character set other than those listed as the client or server character set or setting the language in NLS_LANG on the client will throw one of the following errors:

ORA-12734 ORA-12735 ORA-12736 ORA-12737

With Instant Client Light, the error messages obtained are only in English. Therefore, the valid values for the NLS_LANG setting are of the type:
American_territory.characterset

where, territory can be any valid and supported territory and characterset can be any one the previously listed character sets. Instant Client Light can operate with the OCI environment handles created in the OCI_UTF16 mode.
See Also: Oracle Database Globalization Support Guide for more information about NLS settings.

Operation
To operate in the Instant Client Light mode, an application must set the LD_LIBARARY_PATH environment variable in UNIX systems or the PATH environment variable in Microsoft Windows to a location containing the client and data shared libraries. OCI applications by default look for the OCI Data Shared Library, libociei.so in the LD_LIBRARY_PATH environment variable in UNIX systems or the oraociei11.dll Data Shared Library in the PATH environment variable in Microsoft Windows, to determine if the application should operate in the Instant Client mode. In case this library is not found, then OCI tries to load the Instant
Features Specific to JDBC OCI Driver 6-9

Instant Client Light (English)

Client Light Data Shared Library, libociicus.so in UNIX systems or libociicus10.dll in Microsoft Windows. If this library is found, then the application operates in the Instant Client Light mode. Otherwise, a non-Instant Client mode is assumed.

Installation
Instant Client Light can be installed in one of the following ways:

From OTN You can download the required file from http://www.oracle.com/technology/tech/oci/instantclient/insta ntclient.html For Instant Client Light, instead of downloading and expanding the Basic package, download and unzip the Basic Light package. The instantclient_10_2 directory in which the lightweight libraries are unzipped should be empty before unzipping the files.

From Client Admin Install Instead of copying libociei.so or oraociei11.dll from the ORACLE_HOME/instantclient directory, copy libociicus.so or oraociic10.dll from the ORACLE_HOME/instantclient/light directory. That is, the Instant Client directory on the LD_LIBRARY_PATH environment variable, in UNIX systems, should contain the Instant Client Light Data Shared Library, libociicus.so, instead of the larger OCI Instant Client Data Shared Library, libociei.so. In Microsoft Windows, the PATH environment variable should contain oraociicus10.dll instead of oraociei11.dll.

From Oracle Universal Installer If the Instant Client option is selected from Oracle Universal Installer, then libociei.so (or oraociei11.dll on Microsoft Windows) is installed in the base directory of the installation which is going to be placed on the LD_LIBRARY_PATH environment variable. This is so that Instant Client Light is not enabled by default. The Instant Client Light Data Shared Library, libociicus.so (or oraociicus10.dll on Microsoft Windows), is installed in the light subdirectory of the base directory. Therefore, to operate in the Instant Client Light mode, the OCI Data Shared Library, libociei.so (or oraociei11.dll on Windows) must be deleted or renamed and the Instant Client Light Data Shared Library must be copied from the light subdirectory to the base directory of the installation. For example, if Oracle Universal Installer has installed the Instant Client in my_oraic_11_1 directory on the LD_LIBRARY_PATH environment variable, then one would need to do the following to operate in the Instant Client Light mode:
cd my_oraic_11_1 rm libociei.so mv light/libociicus.so .

Note:

All the Instant Client files should always be copied or installed in an empty directory. This is to ensure that no incompatible binaries exist in the installation.

6-10 Oracle Database JDBC Developer's Guide and Reference

7
Server-Side Internal Driver
This chapter covers the following topics:

Overview of the Server-Side Internal Driver Connecting to the Database Exception-Handling Extensions Session and Transaction Context Testing JDBC on the Server Loading an Application into the Server

Overview of the Server-Side Internal Driver

The server-side internal driver is intrinsically tied to Oracle Database and to the Java Virtual Machine (JVM). The driver runs as part of the same process as the database. It also runs within the default session, the same session in which the JVM was started. The server-side internal driver is optimized to run within the database server and provide direct access to SQL data and PL/SQL subprograms on the local database. The entire JVM operates in the same address space as the database and the SQL engine. Access to the SQL engine is a function call. This enhances the performance of your Java Database Connectivity (JDBC) applications and is much faster than running a remote Oracle Net call to access the SQL engine. The server-side internal driver supports the same features, application programming interfaces (APIs), and Oracle extensions as the client-side drivers. This makes application partitioning very straightforward. For example, if you have a Java application that is data-intensive, then you can easily move it into the database server for better performance, without having to modify the application-specific calls.

Connecting to the Database

As described in the preceding section, the server-side internal driver runs within a default session. Therefore, you are already connected. There are two methods to access the default connection:

Use the OracleDataSource.getConnection method, with any of the following forms as the URL string: jdbc:oracle:kprb jdbc:default:connection jdbc:oracle:kprb:

Server-Side Internal Driver 7-1

Connecting to the Database

jdbc:default:connection:

Use the Oracle-specific defaultConnection method of the OracleDriver class.

Using defaultConnection is generally recommended.

Note:

You are no longer required to register the OracleDriver class for connecting with the server-side internal driver.

Connecting with the OracleDriver Class defaultConnection Method The defaultConnection method of the oracle.jdbc.OracleDriver class is an Oracle extension and always returns the same connection object. Even if you call this method multiple times, assigning the resulting connection object to different variable names, then only a single connection object is reused. You need not include a connection string in the defaultConnection call. For example:
import java.sql.*; import oracle.jdbc.*; class JDBCConnection { public static Connection connect() throws SQLException { Connection conn = null; try { // connect with the server-side internal driver conn = ora.defaultConnection(); } } catch (SQLException e) {...} return conn; } }

Note that there is no conn.close call in the example. When JDBC code is running inside the target server, the connection is an implicit data channel, not an explicit connection instance as from a client. It should not be closed. If you do call the close method, then be aware of the following:

All connection instances obtained through the defaultConnection method, which actually reference the same database connection, will be closed and unavailable for further use, with state and resource cleanup as appropriate. Running defaultConnection afterward would result in a new connection object. Even though the connection object is closed, the implicit connection to the database will not be closed.

Connecting with the OracleDataSource.getConnection Method To connect to the internal server connection from code that is running within the target server, you can use the OracleDataSource.getConnection method with either of the following URLs:
OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:oracle:kprb");

7-2 Oracle Database JDBC Developer's Guide and Reference

Exception-Handling Extensions

Connection conn = ods.getConnection();

or:
OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:default:connection"); Connection conn = ods.getConnection();

or:
OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:oracle:kprb:"); Connection conn = ods.getConnection();

or:
OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:default:connection:"); Connection conn = ods.getConnection();

Any user name or password you include in the URL is ignored in connecting to the default server connection. The OracleDataSource.getConnection method returns a new Java Connection object every time you call it. The fact that OracleDataSource.getConnection returns a new connection object every time you call it is significant if you are working with object maps or type maps. A type map is associated with a specific Connection object and with any state that is part of the object. If you want to use multiple type maps as part of your program, then you can call getConnection to create a new Connection object for each type map.
Note:

Although the OracleDataSource.getConnection method is returning a new object every time you call it, it is not creating a new database connection every time.

Exception-Handling Extensions
The server-side internal driver, in addition to having standard exception-handling capabilities, such as getMessage(), getErrorCode(), and getSQLState(), provides the oracle.jdbc.driver.OracleSQLException class, which is a legacy from the earliest server-side internal driver. This class is a subclass of the standard java.sql.SQLException class and is not available to the client-side JDBC drivers or the server-side Thin driver. When an error condition occurs in the server, it often results in a series of related errors being placed in an internal error stack. The JDBC server-side internal driver retrieves errors from the stack and places them in a chain of OracleSQLException objects. You can use the following methods in processing these exceptions:

SQLException getNextException() This method returns the next exception in the chain or a null value if there are no further exceptions. You can start with the first exception you receive and work through the chain. This is a standard method.

int getNumParameters() (Oracle extension) Errors from the server usually include parameters, or variables, that are part of the error message. These may indicate what type of error occurred, what kind of

Server-Side Internal Driver 7-3

Session and Transaction Context

operation was being attempted, or the invalid or affected values. This method returns the number of parameters included with this error. It is an Oracle extension.

Object[] getParameters() (Oracle extension) This method returns a Java Object[] array containing the parameters included with this error. It is an Oracle extension.

However, in 11g release 1, only a subset of the exceptions thrown by the driver are instances of this class. In 11g release 1 (11.1), this class is deprecated and will be removed in the next release.

Note:

Oracle strongly discourages the use of this class.

Exception-handling differs depending on the version of JDK you are using.

Session and Transaction Context

The server-side driver operates within a default session and default transaction context. The default session is the session in which the JVM was started. In effect, you are already connected to the database on the server. This is different from the client-side where there is no default session. You must explicitly connect to the database. Auto-commit mode is disabled in the server. You must manage transaction COMMIT and ROLLBACK operations explicitly by using the appropriate methods on the connection object:
conn.commit();

or:
conn.rollback();

Note:

As a best practice, it is recommended not to commit or rollback a transaction inside the server.

Testing JDBC on the Server

Almost any JDBC program that can run on a client can also run on the server. All the programs in the samples directory can be run on the server, with only minor modifications. Usually, these modifications concern only the connection statement. Consider the following code fragment which obtains a connection to a database:
ods.setUrl( "jdbc:oracle:oci:@(DESCRIPTION= (ADDRESS=(PROTOCOL=TCP)(HOST=cluster_alias) (PORT=1521)) (CONNECT_DATA=(SERVICE_NAME=service_name)))"); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection();

7-4 Oracle Database JDBC Developer's Guide and Reference

Loading an Application into the Server

We can modify this code fragment for use in the server-side internal driver. In the server-side internal driver, no user, password, or database information is necessary. For the connection statement, you use:
ods.setUrl( "jdbc:oracle:kprb:@"); Connection conn = ods.getConnection();

However, the most convenient way to get a connection is to call the OracleDriver.defaultConnection method, as follows:
Connection conn = OracleDriver.defaultConnection();

Loading an Application into the Server

When loading an application into the server, you can load .class files that you have already compiled on the client or you can load .java source files and have them automatically compiled on the server. In either case, use the loadjava utility to load your files. You can either specify source file names on the command line or put the files into a Java Archive (JAR) file and specify the JAR file name on the command line. The loadjava script, which runs the actual utility, is in the bin directory in your Oracle home. This directory should already be in your path once Oracle has been installed.
Note:

The loadjava utility supports compressed files.

Loading Class Files into the Server Consider a case where you have the following three class files in your application: Foo1.class, Foo2.class, and Foo3.class. Each class is written into its own class schema object in the server. You can load the class files using the default JDBC Oracle Call Interface (OCI) driver in the following ways:

Specifying the individual class file names, as follows:

loadjava -user scott Foo1.class Foo2.class Foo3.class Password: password

Specifying the class file names using a wildcard, as follows:

loadjava -user scott Foo*.class Password: password

Specifying a JAR file that contains the class files, as follows:

loadjava -user scott Foo.jar Password: password

You can load the files using the JDBC Thin driver, as follows:
loadjava -thin -user scott@localhost:1521:ORCL Foo.jar Password: password

Server-Side Internal Driver 7-5

Loading an Application into the Server

Note:

Because the server-side embedded JVM uses Java Development Kit (JDK) 1.5, it is advisable to compile classes under JDK 1.5, if they will be loaded into the server. This will catch incompatibilities during compilation, instead of at run time.

Loading Source Files into the Server If you enable the loadjava -resolve option when loading a .java source file, then the server-side compiler will compile your application as it is loaded, resulting in both a source schema object for the original source code and one or more class schema objects for the compiled output. If you do not specify -resolve, then the source is loaded into a source schema object without any compilation. In this case, however, the source is implicitly compiled the first time an attempt is made to use a class defined in the source. For example, run loadjava as follows to load and compile Foo.java, using the default JDBC OCI driver:
loadjava -user scott -resolve Foo.java Password: password

Or, use the following command to load using the JDBC Thin driver:
loadjava -thin -user scott@localhost:1521:ORCL -resolve Foo.java Password: password

Either of these will result in appropriate class schema objects being created in addition to the source schema object.

Note:

Oracle generally recommends compiling source on the client, whenever possible, and loading the .class files instead of the source files into the server.

7-6 Oracle Database JDBC Developer's Guide and Reference

Part III
Connection and Security
This part consists of chapters that discuss the use of data sources and URLs to connect to the database. It also includes chapters that discuss the security features supported by the Oracle Java Database Connectivity (JDBC) Oracle Call Interface (OCI) and Thin drivers, Secure Sockets Layer (SSL) support in JDBC Thin driver, and middle-tier authentication through proxy connections. Part III contains the following chapters:

Chapter 8, "Data Sources and URLs" Chapter 9, "JDBC Client-Side Security Features" Chapter 10, "Proxy Authentication"

8
Data Sources and URLs
This chapter discusses connecting applications to databases using Java Database Connectivity (JDBC) data sources, as well as the URLs that describe databases. This chapter contains the following sections:

Data Sources Database URLs and Database Specifiers

Data Sources
Data sources are standard, general-use objects for specifying databases or other resources to use. The JDBC 2.0 extension application programming interface (API) introduced the concept of data sources. For convenience and portability, data sources can be bound to Java Naming and Directory Interface (JNDI) entities, so that you can access databases by logical names. The data source facility provides a complete replacement for the previous JDBC DriverManager facility. You can use both facilities in the same application, but it is recommended that you transition your application to data sources. This section covers the following topics:

Overview of Oracle Data Source Support for JNDI Features and Properties of Data Sources Creating a Data Source Instance and Connecting Creating a Data Source Instance, Registering with JNDI, and Connecting Supported Connection Properties Using Roles for SYS Login Configuring Database Remote Login Bequeath Connection and SYS Logon Properties for Oracle Performance Extensions

Overview of Oracle Data Source Support for JNDI

The JNDI standard provides a way for applications to find and access remote services and resources. These services can be any enterprise services. However, for a JDBC application, these services would include database connections and services.

Data Sources and URLs

8-1

Data Sources

JNDI allows an application to use logical names in accessing these services, removing vendor-specific syntax from application code. JNDI has the functionality to associate a logical name with a particular source for a desired service. All Oracle JDBC data sources are JNDI-referenceable. The developer is not required to use this functionality, but accessing databases through JNDI logical names makes the code more portable.
Note:

Using JNDI functionality requires the jndi.jar file to be in the CLASSPATH environment variable. This file is included with the Java products on the installation CD. You must add it to the CLASSPATH environment variable separately. You can also obtain it from the Sun Microsystems Web site, but it is advisable to use the version from Oracle, because it has been tested with the Oracle drivers.

Features and Properties of Data Sources

By using the data source functionality with JNDI, you do not need to register the vendor-specific JDBC driver class name and you can use logical names for URLs and other properties. This ensures that the code for opening database connections is portable to other environments. The DataSource Interface and Oracle Implementation A JDBC data source is an instance of a class that implements the standard javax.sql.DataSource interface:
public interface DataSource { Connection getConnection() throws SQLException; Connection getConnection(String username, String password) throws SQLException; ... }

Oracle implements this interface with the OracleDataSource class in the oracle.jdbc.pool package. The overloaded getConnection method returns a connection to the database. To use other values, you can set properties using appropriate setter methods. For alternative user names and passwords, you can also use the getConnection method that takes these parameters as input. This would take priority over the property settings.
Note: The OracleDataSource class and all subclasses implement the java.io.Serializable and javax.naming.Referenceable interfaces.

Properties of DataSource The OracleDataSource class, as with any class that implements the DataSource interface, provides a set of properties that can be used to specify a database to connect to. These properties follow the JavaBeans design pattern.

8-2 Oracle Database JDBC Developer's Guide and Reference

Data Sources

Table 81 and Table 82 list OracleDataSource properties. The properties in Table 81 are standard properties according to the Sun Microsystems specification. The properties in Table 82 are Oracle extensions.
Note:

Oracle does not implement the standard roleName property.

Table 81 Name

Standard Data Source Properties Type String String Description Name of the particular database on the server. Also known as the SID in Oracle terminology. Name of the underlying data source class. For connection pooling, this is an underlying pooled connection data source class. For distributed transactions, this is an underlying XA data source class. Description of the data source. Network protocol for communicating with the server. For Oracle, this applies only to the JDBC Oracle Call Interface (OCI) drivers and defaults to tcp. Password for the connecting user. Number of the port where the server listens for requests Name of the database server Name for the login

databaseName dataSourceName

description networkProtocol

String String

password portNumber serverName user

String int String String

The OracleDataSource class implements the following setter and getter methods for the standard properties:

public synchronized void setDatabaseName(String dbname) public synchronized String getDatabaseName() public synchronized void setDataSourceName(String dsname) public synchronized String getDataSourceName() public synchronized void setDescription(String desc) public synchronized String getDescription() public synchronized void setNetworkProtocol(String np) public synchronized String getNetworkProtocol() public synchronized void setPassword(String pwd) public synchronized void setPortNumber(int pn) public synchronized int getPortNumber() public synchronized void setServerName(String sn) public synchronized String getServerName() public synchronized void setUser(String user) public synchronized String getUser()
Note:

For security reasons, there is no getPassword() method.

Data Sources and URLs

8-3

Data Sources

Table 82 Name connectionCacheName

Oracle Extended Data Source Properties Type String Description Specifies the name of the cache. This cannot be changed after the cache has been created.

connectionCacheProperties connectionCachingEnabled connectionProperties driverType

java.util.P Specifies properties for implicit connection cache. roperties Boolean Specifies whether implicit connection cache is in use.

java.util.P Specifies the connection properties. roperties String Specifies Oracle JDBC driver type. It can be one of oci, thin, or kprb. Specifies whether Fast Connection Failover is in use. Specifies whether the implicit statement connection cache is enabled. Specifies the maximum time in seconds that this data source will wait while attempting to connect to a database.

fastConnectionFailoverEnabled Boolean implicitCachingEnabled loginTimeout Boolean int

logWriter maxStatements serviceName tnsEntry

java.io.Pri Specifies the log writer for this data source. ntWriter int String String Specifies the maximum number of statements in the application cache. Specifies the database service name for this data source. Specifies the TNS entry name, relevant only for the OCI driver. The TNS entry name corresponds to the TNS entry specified in the tnsnames.ora configuration file. This property is only for OracleXADatasource. Enable this OracleXADataSource property when using the Native XA feature with the OCI driver, to access Oracle pre-8.1.6 databases and later. If the tnsEntry property is not set when using the Native XA feature, then a SQLException with error code ORA-17207 is thrown

url

String

Specifies the URL of the database connection string. Provided as a convenience, it can help you migrate from an older Oracle Database. You can use this property in place of the Oracle tnsEntry and driverType properties and the standard portNumber, networkProtocol, serverName, and databaseName properties. Allows an OracleXADataSource using the Native XA feature with the OCI driver, to access Oracle pre-8.1.6 databases and later. If the nativeXA property is enabled, be sure to set the tnsEntry property as well. This property is only for OracleXADatasource. This DataSource property defaults to false.

nativeXA

Boolean

ONSConfiguration

String

Specifies the ONS configuration string that is used to remotely subscribe to FaN/ONS events.

8-4 Oracle Database JDBC Developer's Guide and Reference

Data Sources

Note:

This table omits properties that supported the deprecated connection cache based on OracleConnectionCache. Because Native XA performs better than Java XA, use Native XA whenever possible.

The OracleDataSource class implements the following setXXX and getXXX methods for the Oracle extended properties:

String getConnectionCacheName() java.util.Properties getConnectionCacheProperties() void setConnectionCacheProperties(java.util.Properties cp) java.util.Properties getConnectionProperties() void setConnectionProperties(java.util.Properties cp)
Note: Use the setConnectionProperties method to set the properties of the connection and the setConnectionCacheProperties method to set the properties of the connection cache.

For more information about the properties of the connection refer to "Supported Connection Properties" on page 8-8. For more information about the properties of the connection refer to "Connection Cache Properties" on page 21-8.

boolean getConnectionCachingEnabled() void setImplicitCachingEnabled() String getDriverType() void setDriverType(String dt) String getURL() void setURL(String url) String getTNSEntryName() void setTNSEntryName(String tns) boolean getNativeXA() void setNativeXA(boolean nativeXA) String getONSConfiguration() void setONSConfiguration(String onsConfig)

If you are using the server-side internal driver, that is, the driverType property is set to kprb, then any other property settings are ignored. If you are using the JDBC Thin or OCI driver, then note the following:

A URL setting can include settings for user and password, as in the following example, in which case this takes precedence over individual user and password property settings:
Data Sources and URLs 8-5

Data Sources

jdbc:oracle:thin:scott/tiger@localhost:1521:orcl

Settings for user and password are required, either directly through the URL setting or through the getConnection call. The user and password settings in a getConnection call take precedence over any property settings. If the url property is set, then any tnsEntry, driverType, portNumber, networkProtocol, serverName, and databaseName property settings are ignored. If the tnsEntry property is set, which presumes the url property is not set, then any databaseName, serverName, portNumber, and networkProtocol settings are ignored. If you are using an OCI driver, which presumes the driverType property is set to oci, and the networkProtocol is set to ipc, then any other property settings are ignored.

Also, note that getConnectionCacheName() will return the name of the cache only if the ConnectionCacheName property of the data source is set after caching is enabled on the data source.

Creating a Data Source Instance and Connecting

This section shows an example of the most basic use of a data source to connect to a database, without using JNDI functionality. Note that this requires vendor-specific, hard-coded property settings. Create an OracleDataSource instance, initialize its connection properties as appropriate, and get a connection instance, as in the following example:
OracleDataSource ods = new OracleDataSource(); ods.setDriverType("oci"); ods.setServerName("dlsun999"); ods.setNetworkProtocol("tcp"); ods.setDatabaseName("816"); ods.setPortNumber(1521); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection();

Or, optionally, override the user name and password, as follows:

Connection conn = ods.getConnection("bill", "lion");

Creating a Data Source Instance, Registering with JNDI, and Connecting

This section exhibits JNDI functionality in using data sources to connect to a database. Vendor-specific, hard-coded property settings are required only in the portion of code that binds a data source instance to a JNDI logical name. From that point onward, you can create portable code by using the logical name in creating data sources from which you will get your connection instances.
Note: Creating and registering data sources is typically handled by a JNDI administrator, not in a JDBC application.

8-6 Oracle Database JDBC Developer's Guide and Reference

Data Sources

Initialize Data Source Properties Create an OracleDataSource instance, and then initialize its properties as appropriate, as in the following example:
OracleDataSource ods = new OracleDataSource(); ods.setDriverType("oci"); ods.setServerName("dlsun999"); ods.setNetworkProtocol("tcp"); ods.setDatabaseName("816"); ods.setPortNumber(1521); ods.setUser("scott"); ods.setPassword("tiger");

Register the Data Source Once you have initialized the connection properties of the OracleDataSource instance ods, as shown in the preceding example, you can register this data source instance with JNDI, as in the following example:
Context ctx = new InitialContext(); ctx.bind("jdbc/sampledb", ods);

Calling the JNDI InitialContext() constructor creates a Java object that references the initial JNDI naming context. System properties, which are not shown, instruct JNDI which service provider to use. The ctx.bind call binds the OracleDataSource instance to a logical JNDI name. This means that anytime after the ctx.bind call, you can use the logical name jdbc/sampledb in opening a connection to the database described by the properties of the OracleDataSource instance ods. The logical name jdbc/sampledb is logically bound to this database. The JNDI namespace has a hierarchy similar to that of a file system. In this example, the JNDI name specifies the subcontext jdbc under the root naming context and specifies the logical name sampledb within the jdbc subcontext. The Context interface and InitialContext class are in the standard javax.naming package.
Note:

The JDBC 2.0 Specification requires that all JDBC data sources be registered in the jdbc naming subcontext of a JNDI namespace or in a child subcontext of the jdbc subcontext.

Open a Connection To perform a lookup and open a connection to the database logically bound to the JNDI name, use the logical JNDI name. Doing this requires casting the lookup result, which is otherwise a Java Object, to OracleDataSource and then using its getConnection method to open the connection. Here is an example:
OracleDataSource odsconn = (OracleDataSource)ctx.lookup("jdbc/sampledb"); Connection conn = odsconn.getConnection();

Data Sources and URLs

8-7

Data Sources

Supported Connection Properties

For a detailed list of connection properties that Oracle JDBC drivers support, see the Javadoc.

Using Roles for SYS Login

To specify the role for the SYS login, use the internal_logon connection property. To logon as SYS, set the internal_logon connection property to SYSDBA or SYSOPER.
Note: The ability to specify a role is supported only for the sys user name.

For a bequeath connection, we can get a connection as SYS by setting the internal_logon property. For a remote connection, we need additional password file setting procedures.

Configuring Database Remote Login

Before the JDBC Thin driver can connect to the database as SYSDBA, you must configure the user, because Oracle Database security system requires a password file for remote connections as an administrator. Perform the following:
1.

Set a password file on the server-side or on the remote database, using the orapwd password utility. You can add a password file for user sys as follows:

In UNIX
orapwd file=$ORACLE_HOME/dbs/orapw entries=200 Enter password: password

In Microsoft Windows
orapwd file=%ORACLE_HOME%\database\PWDsid_name.ora entries=200 Enter password: password

file must be the name of the password file. password is the password for the user SYS. It can be altered using the ALTER USER statement in SQL Plus. You should set entries to a value higher than the number of entries you expect. The syntax for the password file name is different on Microsoft Windows and UNIX.
See Also:
2.

Oracle Database Administrator's Guide

Enable remote login as sysdba. This step grants SYSDBA and SYSOPER system privileges to individual users and lets them connect as themselves. Stop the database, and add the following line to initservice_name.ora, in UNIX, or init.ora, in Microsoft Windows:
remote_login_passwordfile=exclusive

The initservice_name.ora file is located at ORACLE_HOME/dbs/ and also at ORACLE_HOME/admin/db_name/pfile/. Ensure that you keep the two files synchronized. The init.ora file is located at %ORACLE_BASE%\ADMIN\db_name\pfile\.

8-8 Oracle Database JDBC Developer's Guide and Reference

Data Sources

3.

Change the password for the SYS user. This is an optional step.
PASSWORD sys Changing password for sys New password: password Retype new password: password

4.

Verify whether SYS has the SYSDBA privilege.

SQL> select * from v$pwfile_users; USERNAME SYSDB -----------------------------SYS TRUE SYSOP --------TRUE

5.

Restart the remote database.

Example 81 Using SYS Login To Make a Remote Connection //This example works regardless of language settings of the database. /** case of remote connection using sys **/ import java.sql.*; import oracle.jdbc.*; import oracle.jdbc.pool.*; // create an OracleDataSource OracleDataSource ods = new OracleDataSource(); // set connection properties java.util.Properties prop = new java.util.Properties(); prop.put("user", "sys"); prop.put("password", "sys"); prop.put("internal_logon", "sysoper"); ods.setConnectionProperties(prop); // set the url // the url can use oci driver as well as: // url = "jdbc:oracle:oci8:@inst1"; the inst1 is a remote database String url = "jdbc:oracle:thin:@//myHost:1521/service_name"; ods.setURL(url); // get the connection Connection conn = ods.getConnection(); ...

Bequeath Connection and SYS Logon

The following example illustrates how to use the internal_logon and SYSDBA arguments to specify the SYS login. This example works regardless of the database's national-language settings of the database.
/** Example of bequeath connection **/ import java.sql.*; import oracle.jdbc.*; import oracle.jdbc.pool.*; // create an OracleDataSource instance OracleDataSource ods = new OracleDataSource(); // set neccessary properties java.util.Properties prop = new java.util.Properties(); prop.put("user", "sys"); prop.put("password", "sys"); prop.put("internal_logon", "sysdba"); ods.setConnectionProperties(prop);

Data Sources and URLs

8-9

Data Sources

// the url for bequeath connection String url = "jdbc:oracle:oci8:@"; ods.setURL(url); // retrieve the connection Connection conn = ods.getConnection(); ...

Properties for Oracle Performance Extensions

Some of the connection properties are for use with Oracle performance extensions. Setting these properties is equivalent to using corresponding methods on the OracleConnection object, as follows:

Setting the defaultRowPrefetch property is equivalent to calling setDefaultRowPrefetch. Setting the remarksReporting property is equivalent to calling setRemarksReporting.
See Also: "DatabaseMetaData TABLE_REMARKS Reporting" on page 23-19

Setting the defaultBatchValue property is equivalent to calling setDefaultExecuteBatch

See Also:

"Oracle Update Batching" on page 23-3

Example The following example shows how to use the put method of the java.util.Properties class, in this case, to set Oracle performance extension parameters.
//import packages and register the driver import java.sql.*; import java.math.*; import oracle.jdbc.*; import oracle.jdbc.pool.OracleDataSource; //specify the properties object java.util.Properties info = new java.util.Properties(); info.put ("user", "scott"); info.put ("password", "tiger"); info.put ("defaultRowPrefetch","20"); info.put ("defaultBatchValue", "5"); //specify the datasource object OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:oracle:thin:@//myhost:1521/orcl"); ods.setUser("scott"); ods.setPassword("tiger"); ods.setConnectionProperties(info); ...

8-10 Oracle Database JDBC Developer's Guide and Reference

Database URLs and Database Specifiers

Database URLs and Database Specifiers

Database URLs are strings. The complete URL syntax is:
jdbc:oracle:driver_type:[username/password]@database_specifier

Note:

The brackets indicate that the username/password pair is optional. kprb, the internal server-side driver, uses an implicit connection. Database URLs for the server-side driver end after the driver_type.

The first part of the URL specifies which JDBC driver is to be used. The supported driver_type values are thin, oci, and kprb. The remainder of the URL contains an optional user name and password separated by a slash, an @, and the database specifier, which uniquely identifies the database to which the application is connected. Some database specifiers are valid only for the JDBC Thin driver, some only for the JDBC OCI driver, and some for both. Database Specifiers Table 83, shows the possible database specifiers, listing which JDBC drivers support each specifier.
Note:

Starting Oracle Database 10g, Oracle Service IDs are not supported. Oracle Database 10g no longer supports Oracle Names as a naming method.

Table 83 Specifier Oracle Net connection descriptor

Supported Database Specifiers Supported Drivers Thin, OCI Example Thin, using an address list: url="jdbc:oracle:thin:@(DESCRIPTION= (LOAD_BALANCE=on) (ADDRESS_LIST= (ADDRESS=(PROTOCOL=TCP)(HOST=host1) (PORT=1521)) (ADDRESS=(PROTOCOL=TCP)(HOST=host2)(PORT=1521))) (CONNECT_DATA=(SERVICE_NAME=service_name)))" OCI, using a cluster: "jdbc:oracle:oci:@(DESCRIPTION= (ADDRESS=(PROTOCOL=TCP)(HOST=cluster_alias) (PORT=1521)) (CONNECT_DATA=(SERVICE_NAME=service_name)))"

Thin-style service Thin name

Refer to "Thin-style Service Name Syntax" for details. "jdbc:oracle:thin:scott/tiger@//myhost:1521/myservic ename"

Data Sources and URLs 8-11

Database URLs and Database Specifiers

Table 83 (Cont.) Supported Database Specifiers Specifier LDAP syntax Bequeath connection TNSNames alias Supported Drivers Thin OCI Example Refer to LDAP Syntax for details. Empty. That is, nothing after @ "jdbc:oracle:oci:scott/tiger/@" Thin, OCI Refer to "TNSNames Alias Syntax" for details.

Thin-style Service Name Syntax Thin-style service names are supported only by the JDBC Thin driver. The syntax is:
@//host_name:port_number/service_name

For example:
jdbc:oracle:thin:scott/tiger@//myhost:1521/myservicename

Note:

The JDBC Thin driver supports only the TCP/IP protocol.

TNSNames Alias Syntax You can find the available TNSNAMES entries listed in the tnsnames.ora file on the client computer from which you are connecting. On Windows, this file is located in the ORACLE_HOME\NETWORK\ADMIN directory. On UNIX systems, you can find it in the ORACLE_HOME directory or the directory indicated in your TNS_ADMIN environment variable. For example, if you want to connect to the database on host myhost as user scott with password tiger that has a TNSNAMES entry of MyHostString, then write the following:
OracleDataSource ods = new OracleDataSource(); ods.setTNSEntryName("MyTNSAlias"); ods.setUser("scott"); ods.setPassword("tiger"); ods.setDriverType("oci"); Connection conn = ods.getConnection();

The oracle.net.tns_admin system property must be set to the location of the tnsnames.ora file so that the JDBC Thin driver can locate the tnsnames.ora file. For example:
System.setProperty("oracle.net.tns_admin", "c:\\Temp"); String url = "jdbc:oracle:thin:@tns_entry";

Note:

When using TNSNames with the JDBC Thin driver, you must set the oracle.net.tns_admin property to the directory that contains your tnsnames.ora file.

java -Doracle.net.tns_admin=$ORACLE_HOME/network/admin

8-12 Oracle Database JDBC Developer's Guide and Reference

Database URLs and Database Specifiers

LDAP Syntax An example of database specifier using the Lightweight Directory Access Protocol (LDAP) syntax is as follows:
"jdbc:oracle:thin:@ldap://ldap.acme.com:7777/sales,cn=OracleContext,dc=com"

When using SSL, change this to:

"jdbc:oracle:thin:@ldaps://ldap.acme.com:7777/sales,cn=OracleContext,dc=com"

Note:

The JDBC Thin driver can use LDAP over SSL to communicate with Oracle Internet Directory if you substitute ldaps: for ldap: in the database specifier. The LDAP server must be configured to use SSL. If it is not, then the connection attempt will hang.

The JDBC Thin driver supports failover of a list of LDAP servers during the service name resolution process, without the need for a hardware load balancer. Also, client-side load balancing is supported for connecting to LDAP servers. A list of space separated LDAP URLs syntax is used to support failover and load balancing. When a list of LDAP URLs is specified, both failover and load balancing are enabled by default. The oracle.net.ldap_loadbalance connection property can be used to disable load balancing, and the oracle.net.ldap_failover connection property can be used to disable failover. An example, which uses failover, but with client-side load balancing disabled, is as follows:
Properties prop = new Properties(); String url = "jdbc:oracle:thin:@ldap://ldap1.acme.com:3500/cn=salesdept,cn=OracleContext,dc=com /salesdb " + "ldap://ldap2.acme.com:3500/cn=salesdept,cn=OracleContext,dc=com/salesdb " + "ldap://ldap3.acme.com:3500/cn=salesdept,cn=OracleContext,dc=com/salesdb"; prop.put("oracle.net.ldap_loadbalance", "OFF" ); OracleDataSource ods = new OracleDataSource(); ods.setURL(url); ods.setConnectionProperties(prop);

The JDBC Thin driver supports LDAP nonanonymous bind. A set of JNDI environment properties, which contains authentication information, can be specified for a data source. If a LDAP server is configured as not allowing anonymous bind, then authentication information must be provided to connect to the LDAP server. The following example shows a simple clear-text password authentication:
String url = "jdbc:oracle:thin:@ldap://ldap.acme.com:7777/sales,cn=salesdept,cn=OracleContext,d c=com"; Properties prop = new Properties(); prop.put("java.naming.security.authentication", "simple"); prop.put("java.naming.security.principal","cn=salesdept,cn=OracleContext,dc=com"); prop.put("java.naming.security.credentials", "mysecret"); OracleDataSource ods = new OracleDataSource(); ods.setURL(url);

Data Sources and URLs 8-13

Database URLs and Database Specifiers

ods.setConnectionProperties(prop);

Since JDBC passes down the three properties to JNDI, the authentication mechanism chosen by client is consistent with how these properties are interpreted by JNDI. For example, if the client specifies authentication information without explicitly specifying the java.naming.security.authentication property, then the default authentication mechanism is "simple". Please refer to relevant JDNI documentation for details.

8-14 Oracle Database JDBC Developer's Guide and Reference

9
JDBC Client-Side Security Features
This chapter discusses support in the Oracle Java Database Connectivity (JDBC) Oracle Call Interface (OCI) and JDBC Thin drivers for login authentication, data encryption, and data integrity, particularly, with respect to features of the Oracle Advanced Security option. Oracle Advanced Security, previously known as the Advanced Networking Option (ANO) or Advanced Security Option (ASO), provides industry standards-based data encryption, data integrity, third-party authentication, single sign-on, and access authorization. From 11g release 1 (11.1), both the JDBC OCI and Thin drivers support all the Oracle Advanced Security features. Earlier releases of the JDBC drivers did not support some of the ASO features.
Note:

This discussion is not relevant to the server-side internal driver, given that all communication through that driver is completely internal to the server.

This chapter contains the following sections:

Support for Oracle Advanced Security Support for Login Authentication Support for Strong Authentication Support for OS Authentication Support for Data Encryption and Integrity Support for SSL Support for Kerberos Support for RADIUS Secure External Password Store

Support for Oracle Advanced Security

Oracle Advanced Security provides the following security features:

Data Encryption Sensitive information communicated over enterprise networks and the Internet can be protected by using encryption algorithms, which transform information into a form that can be deciphered only with a decryption key. Some of the supported encryption algorithms are RC4, DES, 3DES, and AES.

JDBC Client-Side Security Features

9-1

Support for Oracle Advanced Security

To ensure data integrity during transmission, Oracle Advanced Security generates a cryptographically secure message digest, using MD5 or SHA-1 hashing algorithms, and includes it with each message sent across a network. This protects the communicated data from attacks, such as data modification, deleted packets, and replay attacks.

Strong Authentication To ensure network security in distributed environments, it is necessary to authenticate the user and check his credentials. Password authentication is the most common means of authentication. Oracle Advanced Security enables strong authentication with Oracle authentication adapters, which support various third-party authentication services, including SSL with digital certificates. Oracle Advanced Security supports the following industry-standard authentication methods: Kerberos Remote Authentication Dial-In User Service (RADIUS) Distributed Computing Environment (DCE) Secure Sockets Layer (SSL)
Tip:

Oracle Database Advanced Security Administrator's Guide

JDBC OCI Driver Support for Oracle Advanced Security If you are using the JDBC OCI driver, which presumes you are running from a computer with an Oracle client installation, then support for Oracle Advanced Security and incorporated third-party features is fairly similar to the support provided by in any Oracle client situation. Your use of Advanced Security features is determined by related settings in the sqlnet.ora file on the client computer. Starting from Oracle Database 11g Release 1 (11.1), the JDBC OCI driver attempts to use external authentication if you try connecting to a database without providing a password. The following are some examples using the JDBC OCI driver to connect to a database without providing a password: SSL Authentication Example 91 uses SSL authentication to connect to the database.
Example 91 import java.sql.*; import java.util.Properties; public class test { public static void main( String [] args ) throws Exception { String url = "jdbc:oracle:oci:@" +"(DESCRIPTION=(ADDRESS=(PROTOCOL=tcps)(HOST=stadh25)(PORT=1529))" +"(CONNECT_DATA=(SERVICE_NAME=mydatabaseinstance)))"; Driver driver = new oracle.jdbc.OracleDriver(); Properties props = new Properties(); Connection conn = driver.connect( url, props ); conn.close(); } }

9-2 Oracle Database JDBC Developer's Guide and Reference

Support for Strong Authentication

Using Data Source Example 92 uses a data source to connect to the database.
Example 92 import import import import java.sql.*; javax.sql.*; java.util.Properties; oracle.jdbc.pool.*;

public class testpool { public static void main( String args ) throws Exception { String url = "jdbc:oracle:oci:@" +"(DESCRIPTION=(ADDRESS=(PROTOCOL=tcps)(HOST=stadh25)(PORT=1529))" +"(CONNECT_DATA=(SERVICE_NAME=mydatabaseinstance)))"; OracleConnectionPoolDataSource ocpds = new OracleConnectionPoolDataSource(); ocpds.setURL(url); PooledConnection pc = ocpds.getPooledConnection(); Connection conn = pc.getConnection(); } }

Note:

The key exception to the preceding, with respect to Java, is that the Secure Sockets Layer (SSL) protocol is supported by the Oracle JDBC OCI drivers only if you use native threads in your application. This requires special attention, because green threads are generally the default.

JDBC Thin Driver Support for Oracle Advanced Security The JDBC Thin driver cannot assume the existence of an Oracle client installation or the presence of the sqlnet.ora file. Therefore, it uses a Java approach to support Oracle Advanced Security. Java classes that implement Oracle Advanced Security are included in the ojdbc5.jar and ojdbc6.jar files. Security parameters for encryption and integrity, usually set in sqlnet.ora, are set using a Java Properties object or through system properties.

Support for Login Authentication

Basic login authentication through JDBC consists of user names and passwords, as with any other means of logging in to an Oracle server. Specify the user name and password through a Java properties object or directly through the getConnection method call. This applies regardless of which client-side Oracle JDBC driver you are using, but is irrelevant if you are using the server-side internal driver, which uses a special direct connection and does not require a user name or password. Starting with 11g release 1 (11.1), the Oracle JDBC Thin driver implements Oracle O5Logon challenge-response protocol instead of O3Logon to authenticate the user.

Support for Strong Authentication

Oracle Advanced Security enables Oracle Database users to authenticate externally. External authentication can be with RADIUS, KERBEROS, Certificate-Based Authentication, Token Cards, Smart Cards, and DCE. This is called strong

JDBC Client-Side Security Features

9-3

Support for OS Authentication

authentication. Oracle JDBC drivers provide support for the following strong authentication methods:

Kerberos RADIUS SSL (certificate-based authentication)

Support for OS Authentication

Operating System (OS) authentication allows Oracle to pass control of user authentication to the operating system. It allows the users to connect to the database by authenticating their OS username in the database. No password is associated with the account since it is assumed that OS authentication is sufficient. In other words, the server delegates the authentication to the client OS. You need to perform the following steps to achieve this:

Use the following command to check the value of the Oracle OS_AUTHENT_PREFIX initialization parameter:
SQL> SHOW PARAMETER os_authent_prefix NAME TYPE VALUE ------------------------------------ ----------- -----------------------------os_authent_prefix string ops$ SQL>

Note:

Remember the OS authentication prefix. You need to create a database user to allow an OS authenticated connection, where the username must be the prefix value concatenated to the OS username.

Add the following line in the t_init1.ora file:

REMOTE_OS_AUTHENT = TRUE

When a connection is attempted from the local database server, the OS username is passed to the Oracle server. If the username is recognized, the Oracle the connection is accepted, otherwise the connection is rejected.

Configuration Steps for Linux

The configuration steps necessary to set up OS authentication on Linux are the following:
1.

Use the following commands to create an OS user w_rose:

# useradd w_rose # passwd w_rose Changing password for w_rose New password: password Retype new password: password

2.

Use the following command to create a database user to allow an OS authenticated connection:
CREATE USER ops$w_rose IDENTIFIED EXTERNALLY; GRANT CONNECT TO ops$w_rose;

3.

Use the following commands to test the OS authentication connection:

9-4 Oracle Database JDBC Developer's Guide and Reference

Support for OS Authentication

su - w_rose export ORACLE_HOME=/u01/app/oracle/product/10.1.0/db_1 export PATH=$PATH:$ORACLE_HOME/bin export ORACLE_SID=DEV1 sqlplus / SQL*Plus: Release 10.1.0.3.0 - Production on Wed Jun 7 08:41:15 2006 Copyright (c) 1982, 2004, Oracle. All rights reserved.

Connected to: Oracle Database 10g Enterprise Edition Release 10.1.0.3.0 - Production With the Partitioning, Oracle Label Security, OLAP and Data Mining options SQL>

Configuration Steps for Windows

The configuration steps necessary to set up OS authentication on Windows are the following:
1.

Create a local user, say, w_rose, using the Computer Management dialog box. For this you have to do the following:
1. 2. 3. 4. 5. 6.

Click Start. From the Start menu, select Programs, then select Administrative Tools and then select Computer Management. Expand Local Users and Groups by clicking on the Plus ("+") sign. Click Users. Select New User from the Action menu. Enter details of the user in the New User dialog box and click Create.
Note:

The preceding steps are only for creating a local user. Domain users can be created in Active Directory.

2.

Use the following command to create a database user to allow an OS authenticated connection:
CREATE USER "OPS$yourdomain.com\p_floyd" IDENTIFIED EXTERNALLY; GRANT CONNECT TO "OPS$yourdomain.com\p_floyd";

Note:

When you create the database user in Windows environment, the user name should be in the following format:

<OS_authentication_prefix_parameter>$<DOMAIN>\<OS_user_name>

When using a Windows server, there is an additional consideration. The following option must be set in the %ORACLE_HOME%\network\admin\sqlnet.ora file:
SQLNET.AUTHENTICATION_SERVICES= (NTS) 3.

Use the following commands to test the OS authentication connection:

JDBC Client-Side Security Features 9-5

Support for Data Encryption and Integrity

C:\> set ORACLE_SID=DB10G C:\> sqlplus / SQL*Plus: Release 10.2.0.1.0 - Production on Tue Oct 17 11:47:01 2006 Copyright (c) 1982, 2005, Oracle. All rights reserved. Connected to: Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - Production With the Partitioning, OLAP and Data Mining options SQL>

JDBC Code Using OS Authentication

Now that you have set up OS authentication to connect to the database, you can use the following JDBC code for connecting to the database:
String url = "jdbc:oracle:thin:@oracleserver.mydomain.com:5521:dbja" Driver driver = new oracle.jdbc.OracleDriver(); DriverManager.registerDriver(driver); Properties props = new Properties(); Connection conn = DriverManager.getConnection( url, props);

The preceding code assumes that it is executed by p_floyd on the client machine. The JDBC drivers retrieve the OS username from the user.name system property that is set by the JVM. As a result, the following thin driver-specific error no longer exists:
ORA-17443=Null user or password not supported in THIN driver

Note:

By default, the JDBC driver retrieves the OS username from the user.name system property, which is set by the JVM. If the JDBC driver is unable to retrieve this system property or if you want to override the value of this system property, then you can use the OracleConnection.CONNECTION_PROPERTY_THIN_VSESSION_O SUSER connection property. For more information, see Oracle Javadoc.

Support for Data Encryption and Integrity

You can use Oracle Advanced Security data encryption and integrity features in your Java database applications, depending on related settings in the server. When using the JDBC OCI driver, set parameters as you would in any Oracle client situation. When using the Thin driver, set parameters through a Java properties object. Encryption is enabled or disabled based on a combination of the client-side encryption-level setting and the server-side encryption-level setting. Similarly, integrity is enabled or disabled based on a combination of the client-side integrity-level setting and the server-side integrity-level setting. Encryption and integrity support the same setting levels, REJECTED, ACCEPTED, REQUESTED, and REQUIRED. Table 91 shows how these possible settings on the client-side and server-side combine to either enable or disable the feature. By default, remote OS authentication (through TCP) is disabled in the database for obvious security reasons.

9-6 Oracle Database JDBC Developer's Guide and Reference

Support for Data Encryption and Integrity

Table 91

Client/Server Negotiations for Encryption or Integrity Client Rejected Client Accepted (default) OFF OFF ON ON Client Requested OFF ON ON ON Client Required connection fails ON ON ON

Server Rejected Server Accepted (default) Server Requested Server Required

OFF OFF OFF connection fails

Table 91 shows, for example, that if encryption is requested by the client, but rejected by the server, it is disabled. The same is true for integrity. As another example, if encryption is accepted by the client and requested by the server, it is enabled. And, again, the same is true for integrity.
See Also:

Oracle Database Advanced Security Administrator's Guide

Note:

The term checksum still appears in integrity parameter names, but is no longer used otherwise. For all intents and purposes, checksum and integrity are synonymous.

This section covers the following topics:

JDBC OCI Driver Support for Encryption and Integrity JDBC Thin Driver Support for Encryption and Integrity Setting Encryption and Integrity Parameters in Java

JDBC OCI Driver Support for Encryption and Integrity

If you are using the JDBC OCI driver, which presumes an Oracle-client setting with an Oracle client installation, then you can enable or disable data encryption or integrity and set related parameters as you would in any Oracle client situation, through settings in the SQLNET.ORA file on the client. To summarize, the client parameters are shown in Table 92:

JDBC Client-Side Security Features

9-7

Support for Data Encryption and Integrity

Table 92

OCI Driver Client Parameters for Encryption and Integrity Parameter Name SQLNET.ENCRYPTION_CLIENT Possible Settings REJECTED ACCEPTED REQUESTED REQUIRED RC4_40, RC4_56, DES, DES40, AES128, AES192, AES256, 3DES112, 3DES168 (see Note)

Parameter Description Client encryption level

Client encryption selected list

SQLNET.ENCRYPTION_TYPES_CLIENT

Client integrity level

SQLNET.CRYPTO_CHECKSUM_CLIENT

REJECTED ACCEPTED REQUESTED REQUIRED MD5, SHA-1

Client integrity selected list

SQLNET.CRYPTO_CHECKSUM_TYPES_CLIENT

Note:

For the Oracle Advanced Security domestic edition only, settings of RC4_128 and RC4_256 are also possible. Oracle Database Advanced Security Administrator's Guide

See Also:

JDBC Thin Driver Support for Encryption and Integrity

The JDBC Thin driver support for data encryption and integrity parameter settings parallels the JDBC OCI driver support discussed in the preceding section. Corresponding parameters can be set through a Java properties object that you would then be used when opening a database connection. Table 93 lists the parameter information for the JDBC Thin driver. These parameters are defined in the oracle.jdbc.OracleConnection interface.
Table 93 Thin Driver Client Parameters for Encryption and Integrity Parameter Type Possible Settings REJECTED ACCEPTED REQUESTED REQUIRED AES256, AES192, AES128, 3DES168, 3DES112, DES56C, DES40C, RC4_256, RC4_128, RC4_40, RC4_56 REJECTED ACCEPTED REQUESTED REQUIRED MD5, SHA1

Parameter Name

CONNECTION_PROPERTY_THIN_NET_ENCRYPTION_LEVEL String

CONNECTION_PROPERTY_THIN_NET_ENCRYPTION_TYPES String

CONNECTION_PROPERTY_THIN_NET_CHECKSUM_LEVEL

String

CONNECTION_PROPERTY_THIN_NET_CHECKSUM_TYPES

String

9-8 Oracle Database JDBC Developer's Guide and Reference

Support for Data Encryption and Integrity

Note:

Because Oracle Advanced Security support for the Thin driver is incorporated directly into the JDBC classes JAR file, there is only one version, not separate domestic and export editions. Only parameter settings that would be suitable for an export edition are possible. The letter C in DES40C and DES56C refers to Cipher Block Chaining (CBC) mode.

Setting Encryption and Integrity Parameters in Java

Use a Java properties object, that is, an instance of java.util.Properties, to set the data encryption and integrity parameters supported by the JDBC Thin driver. The following example instantiates a Java properties object, uses it to set each of the parameters in Table 93, and then uses the properties object in opening a connection to the database:
... Properties prop = new Properties(); prop.setProperty(OracleConnection.CONNECTION_PROPERTY_THIN_NET_ENCRYPTION_LEVEL, "REQUIRED"); prop.setProperty(OracleConnection.CONNECTION_PROPERTY_THIN_NET_ENCRYPTION_TYPES, "( DES40C )"); prop.setProperty(OracleConnection.CONNECTION_PROPERTY_THIN_NET_CHECKSUM_LEVEL, "REQUESTED"); prop.setProperty(OracleConnection.CONNECTION_PROPERTY_THIN_NET_CHECKSUM_TYPES, "( MD5 )"); OracleDataSource ods = new OracleDataSource(); ods.setProperties(prop); ods.setURL("jdbc:oracle:thin:@localhost:1521:main"); Connection conn = ods.getConnection(); ...

The parentheses around the values encryption type and checksum type allow for lists of values. When multiple values are supplied, the server and the client negotiate to determine which value is to be actually used. Example Example 93 is a complete class that sets data encryption and integrity parameters before connecting to a database to perform a query.
Note:

In the example, the string "REQUIRED" is retrieved dynamically through functionality of the AnoServices and Service classes. You have the option of retrieving the strings in this manner or hardcoding them as in the previous examples

Before running this example, you must turn on encryption in the sqlnet.ora file. For example, the following lines will turn on AES256, AES192, and AES128 for the encryption and MD5 and SHA1 for the checksum:
SQLNET.ENCRYPTION_SERVER = ACCEPTED SQLNET.CRYPTO_CHECKSUM_SERVER = ACCEPTED SQLNET.CRYPTO_CHECKSUM_TYPES_SERVER= (MD5, SHA1)

JDBC Client-Side Security Features

9-9

Support for Data Encryption and Integrity

SQLNET.ENCRYPTION_TYPES_SERVER= (AES256, AES192, AES128) SQLNET.CRYPTO_SEED = 2z0hslkdharUJCFtkwbjOLbgwsj7vkqt3bGoUylihnvkhgkdsbdskkKGhdk Example 93 Setting Data Encryption and Integrity Parameters import import import import java.sql.*; java.util.Properties; oracle.net.ano.AnoServices; oracle.jdbc.*;

public class DemoAESAndSHA1 { static final String USERNAME= "scott"; static final String PASSWORD= "tiger"; static final String URL = "jdbc:oracle:thin:@(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=WXYZ)(PORT=5561))" +"(CONNECT_DATA=(SERVICE_NAME=mydatabaseinstance)))"; public static final void main(String[] argv) { DemoAESAndSHA1 demo = new DemoAESAndSHA1(); try { demo.run(); }catch(SQLException ex) { ex.printStackTrace(); } } void run() throws SQLException { OracleDriver dr = new OracleDriver(); Properties prop = new Properties(); // We require the connection to be encrypted with either AES256 or AES192. // If the database doesn't accept such a security level, then the connection attempt will fail.

prop.setProperty(OracleConnection.CONNECTION_PROPERTY_THIN_NET_ENCRYPTION_LEVEL,An oServices.ANO_REQUIRED); prop.setProperty(OracleConnection.CONNECTION_PROPERTY_THIN_NET_ENCRYPTION_TYPES,"( " + AnoServices.ENCRYPTION_AES256 + "," + AnoServices.ENCRYPTION_AES192 + ")"); // We also require the use of the SHA1 algorithm for data integrity checking.

prop.setProperty(OracleConnection.CONNECTION_PROPERTY_THIN_NET_CHECKSUM_LEVEL,AnoS ervices.ANO_REQUIRED); prop.setProperty(OracleConnection.CONNECTION_PROPERTY_THIN_NET_CHECKSUM_TYPES,"( " + AnoServices.CHECKSUM_SHA1 + " )"); prop.setProperty("user",DemoAESAndSHA1.USERNAME); prop.setProperty("password",DemoAESAndSHA1.PASSWORD); OracleConnection oraConn = (OracleConnection)dr.connect(DemoAESAndSHA1.URL,prop); System.out.println("Connection created! Encryption algorithm is: " +

9-10 Oracle Database JDBC Developer's Guide and Reference

Support for SSL

oraConn.getEncryptionAlgorithmName() + ", data integrity algorithm is: " + oraConn.getDataIntegrityAlgorithmName()); oraConn.close(); } }

Support for SSL

Oracle Database 10g provides support for the Secure Sockets Layer (SSL) protocol. SSL is a widely used industry standard protocol that provides secure communication over a network. SSL provides authentication, data encryption, and data integrity. It provides a secure enhancement to the standard TCP/IP protocol, which is used for Internet communication.. SSL uses digital certificates that comply with the X.509v3 standard for authentication and a public and private key pair for encryption. SSL also uses secret key cryptography and digital signatures to ensure privacy and integrity of data. When a network connection over SSL is initiated, the client and server perform an SSL handshake that includes the following steps:

Client and server negotiate about the cipher suites to use. This includes deciding on the encryption algorithms to be used for data transfer. Server sends its certificate to the client, and the client verifies that the certificate was signed by a trusted certification authority (CA). This step verifies the identity of the server. If client authentication is required, the client sends its own certificate to the server, and the server verifies that the certificate was signed by a trusted CA. Client and server exchange key information using public key cryptography. Based on this information, each generates a session key. All subsequent communications between the client and the server is encrypted and decrypted by using this set of session keys and the negotiated cipher suite.
Note: In Oracle Database 11g Release 1 (11.1), SSL authentication is supported in the thin driver. So, you do not need to provide a username/password pair if you are using SSL authentication.

SSL Terminology The following terms are commonly used in the SSL context:

certificate: A certificate is a digitally signed document that binds a public key with an entity. The certificate can be used to verify that the public key belongs to that individual. certification authority: A certification authority (CA), also known as certificate authority, is an entity which issues digitally signed certificates for use by other parties. cipher suite: A cipher suite is a set of cryptographic algorithms and key sizes used to encrypt data sent over an SSL-enabled network. private key: A private key is a secret key, which is never transmitted over a network. The private key is used to decrypt a message that has been encrypted using the corresponding public key. It is also used to sign certificates. The certificate is verified using the corresponding public key.
JDBC Client-Side Security Features 9-11

Support for SSL

public key: A public key is an encryption key that can be made public or sent by ordinary means such as an e-mail message. The public key is used for encrypting the message sent over SSL. It is also used to verify a certificate signed by the corresponding private key. wallet: A wallet is a password-protected container that is used to store authentication and signing credentials, including private keys, certificates, and trusted certificates required by SSL.

Java Version of SSL The Java Secure Socket Extension (JSSE) provides a framework and an implementation for a Java version of the SSL and TLS protocols. JSSE provides support for data encryption, server and client authentication, and message integrity. It abstracts the complex security algorithms and handshaking mechanisms and simplifies application development by providing a building block for application developers, which they can directly integrate into their applications. JSSE is integrated into Java Development Kit (JDK) 1.4 and later, and supports SSL version 2.0 and 3.0. Oracle strongly recommends that you have a clear understanding of the JavaTM Secure Socket Extension (JSSE) framework by Sun Microsystems before using SSL in the Oracle JDBC drivers. The JSSE standard application programming interface (API) is available in the javax.net, javax.net.ssl, and javax.security.cert packages. These packages provide classes for creating and configuring sockets, server sockets, SSL sockets, and SSL server sockets. The packages also provide a class for secure HTTP connections, a public key certificate API compatible with JDK1.1-based platforms, and interfaces for key and trust managers. SSL works the same way, as in any networking environment, in Oracle Database 10g. This section covers the following:

Managing Certificates and Wallets Keys and certificates containers

Managing Certificates and Wallets

To establish an SSL connection with a JDBC client, Thin or OCI, Oracle database server sends its certificate, which is stored in its wallet. The client may or may not need a certificate or wallet depending on the server configuration. The Oracle JDBC Thin driver uses the JSSE framework to create an SSL connection. It uses the default provider (SunJSSE) to create an SSL context. However you can provide your own provider. You do not need a certificate for the client, unless the SSL_CLIENT_AUTHENTICATION parameter is set on the server.

Keys and certificates containers

Java clients can use multiple types of containers such as Oracle wallets, JKS, PKCS12, and so on, as long as a provider is available. For Oracle wallets, OraclePKI provider must be used because the PKCS12 support provided by SunJSSE provider does not support all the features of PKCS12. In order to use OraclePKI provider, the following JARs are required:

oraclepki.jar osdt_cert.jar

9-12 Oracle Database JDBC Developer's Guide and Reference

Support for Kerberos

osdt_core.jar

All these JAR files should be under $ORACLE_HOME/jlib directory.

Support for Kerberos

Oracle Database 11g Release 1 (11.1) introduces support for Kerberos. Kerberos is a network authentication protocol that provides the tools of authentication and strong cryptography over the network. Kerberos helps you secure your information systems across your entire enterprise by using secret-key cryptography. The Kerberos protocol uses strong cryptography so that a client or a server can prove its identity to its server or client across an insecure network connection. After a client and server have used Kerberos to prove their identity, they can also encrypt all of their communications to assure privacy and data integrity as they go about their business. The Kerberos architecture is centered around a trusted authentication service called the key distribution center, or KDC. Users and services in a Kerberos environment are referred to as principals; each principal shares a secret, such as a password, with the KDC. A principal can be a user such as scott or a database server instance.

Configuring Oracle Database to Use Kerberos

Perform the following steps to configure Oracle Database to use Kerberos:
1.

Use the following command to connect to the database:

SQL> connect system Enter password: password

2.

Use the following commands to create a user CLIENT@US.ORACLE.COM that is identified externally:
SQL> create user "CLIENT@US.ORACLE.COM" identified externally; SQL> grant create session to "CLIENT@US.ORACLE.COM";

3.

Use the following commands to connect to the database as sysdba and dismount it:
SQL> connect / as sysdba SQL> shutdown immediate;

4.

Add the following line to $T_WORK/t_init1.ora file:

OS_AUTHENT_PREFIX=""

5.

Use the following command to restart the database:

SQL> startup pfile=t_init1.ora

6.

Modify the sqlnet.ora file to include the following lines:

names.directory_path = (tnsnames) #Kerberos sqlnet.authentication_services = (beq,kerberos5) sqlnet.authentication_kerberos5_service = dbji sqlnet.kerberos5_conf = /home/Jdbc/Security/kerberos/krb5.conf sqlnet.kerberos5_keytab = /home/Jdbc/Security/kerberos/dbji.oracleserver sqlnet.kerberos5_conf_mit = true sqlnet.kerberos_cc_name = /tmp/krb5cc_5088 # logging (optional):

JDBC Client-Side Security Features 9-13

Support for Kerberos

trace_level_server=16 trace_directory_server=/scratch/sqlnet/ 7.

Use the following commands to verify that you can connect through SQL*Plus:
> kinit client > klist Ticket cache: FILE:/tmp/krb5cc_5088 Default principal: client@US.ORACLE.COM Valid starting 06/22/06 07:13:29 Expires 06/22/06 17:13:29 Service principal krbtgt/US.ORACLE.COM@US.ORACLE.COM

Kerberos 4 ticket cache: /tmp/tkt5088 klist: You have no tickets cached > sqlplus '/@(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=oracleserver.mydomain.com)(PORT=55 29)) (CONNECT_DATA=(SERVICE_NAME=mydatabaseinstance)))'

Code Example
This following example demonstrates the new Kerberos authentication feature that is part of Oracle Database 11g Release 1 (11.1) JDBC thin driver. This demo covers two scenarios:

In the first scenario, the OS maintains the user name and credentials. The credentials are stored in the cache and the driver retrieves the credentials before trying to authenticate to the server. This scenario is in the module connectWithDefaultUser().
Note:
1.Before you run this part of the demo, use the following command to verify that you have valid credentials:

> /usr/kerberos/bin/kinit client where, the password is welcome.

2.

Use the following command to list your tickets: > /usr/kerberos/bin/klist

The second scenario covers the case where the application wants to control the user credentials. This is the case of the application server where multiple web users have their own credentials. This scenario is in the module connectWithSpecificUser().
Note:

To run this demo, you need to have a working setup, that is, a Kerberos server up and running, and an Oracle database server that is configured to use Kerberos authentication. You then need to change the URLs used in the example to compile and run it.

Example 94 import com.sun.security.auth.module.Krb5LoginModule; import java.io.IOException;

9-14 Oracle Database JDBC Developer's Guide and Reference

Support for Kerberos

import import import import import import import import import import import import

java.security.PrivilegedExceptionAction; java.sql.Connection; java.sql.ResultSet; java.sql.SQLException; java.sql.Statement; java.util.HashMap; java.util.Properties; javax.security.auth.Subject; javax.security.auth.callback.Callback; javax.security.auth.callback.CallbackHandler; javax.security.auth.callback.PasswordCallback; javax.security.auth.callback.UnsupportedCallbackException;

import oracle.jdbc.OracleConnection; import oracle.jdbc.OracleDriver; import oracle.net.ano.AnoServices; public class KerberosJdbcDemo { String url ="jdbc:oracle:thin:@(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)"+ "(HOST=oracleserver.mydomain.com)(PORT=5561))(CONNECT_DATA=" + "(SERVICE_NAME=mydatabaseinstance)))"; public static void main(String[] arv) { /* If you see the following error message [Mechanism level: Could not load * configuration file c:\winnt\krb5.ini (The system cannot find the path * specified] it's because the JVM cannot locate your kerberos config file. * You have to provide the location of the file. For example, on Windows, * the MIT Kerberos client uses the config file: C\WINDOWS\krb5.ini: */ // System.setProperty("java.security.krb5.conf","C:\\WINDOWS\\krb5.ini"); System.setProperty("java.security.krb5.conf","/home/Jdbc/Security/kerberos/krb5.co nf"); KerberosJdbcDemo kerberosDemo = new KerberosJdbcDemo(); try { System.out.println("Attempt to connect with the default user:"); kerberosDemo.connectWithDefaultUser(); } catch (Exception e) { e.printStackTrace(); } try { System.out.println("Attempt to connect with a specific user:"); kerberosDemo.connectWithSpecificUser(); } catch (Exception e) { e.printStackTrace(); } }

void connectWithDefaultUser() throws SQLException {

JDBC Client-Side Security Features 9-15

Support for Kerberos

OracleDriver driver = new OracleDriver(); Properties prop = new Properties();

prop.setProperty(OracleConnection.CONNECTION_PROPERTY_THIN_NET_AUTHENTICATION_SERV ICES, "("+AnoServices.AUTHENTICATION_KERBEROS5+")"); prop.setProperty(OracleConnection.CONNECTION_PROPERTY_THIN_NET_AUTHENTICATION_KRB5 _MUTUAL, "true"); /* If you get the following error [Unable to obtain Princpal Name for * authentication] although you know that you have the right TGT in your * credential cache, then it's probably because the JVM can't locate your * cache. * * Note that the default location on windows is "C:\Documents and Settings\krb5cc_username". */ // prop.setProperty(OracleConnection.CONNECTION_PROPERTY_THIN_NET_AUTHENTICATION_KRB5 _CC_NAME, /* On linux: > which kinit /usr/kerberos/bin/kinit > ls -l /etc/krb5.conf lrwxrwxrwx 1 root root 47 Jun 22 06:56 /etc/krb5.conf -> /home/Jdbc/Security/kerberos/krb5.conf > kinit client Password for client@US.ORACLE.COM: > klist Ticket cache: FILE:/tmp/krb5cc_5088 Default principal: client@US.ORACLE.COM Valid starting 11/02/06 09:25:11 Expires 11/02/06 19:25:11 Service principal krbtgt/US.ORACLE.COM@US.ORACLE.COM

Kerberos 4 ticket cache: /tmp/tkt5088 klist: You have no tickets cached */ prop.setProperty(OracleConnection.CONNECTION_PROPERTY_THIN_NET_AUTHENTICATION_KRB5 _CC_NAME, "/tmp/krb5cc_5088"); Connection conn = driver.connect(url,prop); String auth = ((OracleConnection)conn).getAuthenticationAdaptorName(); System.out.println("Authentication adaptor="+auth); printUserName(conn); conn.close(); }

void connectWithSpecificUser() throws Exception { Subject specificSubject = new Subject();

9-16 Oracle Database JDBC Developer's Guide and Reference

Support for Kerberos

// This first part isn't really meaningful to the sake of this demo. In // a real world scenario, you have a valid "specificSubject" Subject that // represents a web user that has valid Kerberos credentials. Krb5LoginModule krb5Module = new Krb5LoginModule(); HashMap sharedState = new HashMap(); HashMap options = new HashMap(); options.put("doNotPrompt","false"); options.put("useTicketCache","false"); options.put("principal","client@US.ORACLE.COM");

krb5Module.initialize(specificSubject,newKrbCallbackHandler(),sharedState,options) ; boolean retLogin = krb5Module.login(); krb5Module.commit(); if(!retLogin) throw new Exception("Kerberos5 adaptor couldn't retrieve credentials (TGT) from the cache"); // to use the TGT from the cache: // options.put("useTicketCache","true"); // options.put("doNotPrompt","true"); // options.put("ticketCache","C:\\Documents and Settings\\Jean de Lavarene\\krb5cc"); // krb5Module.initialize(specificSubject,null,sharedState,options);

// Now we have a valid Subject with Kerberos credentials. The second scenario // really starts here: // execute driver.connect(...) on behalf of the Subject 'specificSubject': Connection conn = (Connection)Subject.doAs(specificSubject, new PrivilegedExceptionAction() { public Object run() { Connection con = null; Properties prop = new Properties(); prop.setProperty(AnoServices.AUTHENTICATION_PROPERTY_SERVICES, "(" + AnoServices.AUTHENTICATION_KERBEROS5 + ")"); try { OracleDriver driver = new OracleDriver(); con = driver.connect(url, prop); } catch (Exception except) { except.printStackTrace(); } return con; } }); String auth = ((OracleConnection)conn).getAuthenticationAdaptorName(); System.out.println("Authentication adaptor="+auth); printUserName(conn); conn.close(); } void printUserName(Connection conn) throws SQLException

JDBC Client-Side Security Features 9-17

Support for RADIUS

{ Statement stmt = null; try { stmt = conn.createStatement(); ResultSet rs = stmt.executeQuery("select user from dual"); while(rs.next()) System.out.println("User is:"+rs.getString(1)); rs.close(); } finally { if(stmt != null) stmt.close(); } } } class KrbCallbackHandler implements CallbackHandler { public void handle(Callback[] callbacks) throws IOException, UnsupportedCallbackException { for (int i = 0; i < callbacks.length; i++) { if (callbacks[i] instanceof PasswordCallback) { PasswordCallback pc = (PasswordCallback)callbacks[i]; System.out.println("set password to 'welcome'"); pc.setPassword((new String("welcome")).toCharArray()); } else { throw new UnsupportedCallbackException(callbacks[i], "Unrecognized Callback"); } } } }

Support for RADIUS

Oracle Database 11g Release 1 (11.1) introduces support for Remote Authentication Dial-In User Service (RADIUS). RADIUS is a client/server security protocol that is most widely known for enabling remote authentication and access. Oracle Advanced Security uses this standard in a client/server network environment to enable use of any authentication method that supports the RADIUS protocol. RADIUS can be used with a variety of authentication mechanisms, including token cards and smart cards. This section contains the following sections:

Configuring Oracle Database to Use RADIUS Code Example

Configuring Oracle Database to Use RADIUS

Perform the following steps to configure Oracle Database to use RADIUS:
1.

Use the following command to connect to the database:

9-18 Oracle Database JDBC Developer's Guide and Reference

Support for RADIUS

SQL> connect system Enter password: password 2.

Use the following commands to create a new user aso from within a database:
SQL> create user aso identified externally; SQL> grant create session to aso;

3.

Use the following commands to connect to the database as sysdba and dismount it:
SQL> connect / as sysdba SQL> shutdown immediate;

4.

Add the following lines to the t_init1.ora file:

os_authent_prefix = ""

Note:

Once the test is over, you need to revert the preceding changes made to the t_init1.ora file.

5.

Use the following command to restart the database:

SQL> startup pfile=?/work/t_init1.ora

6.

Modify the sqlnet.ora file so that it contains only these lines:

sqlnet.authentication_services = ( beq, radius) sqlnet.radius_authentication = <RADUIUS_SERVER_HOST_NAME> sqlnet.radius_authentication_port = 1812 sqlnet.radius_authentication_timeout = 120 sqlnet.radius_secret=/home/Jdbc/Security/radius/radius_key # logging (optional): trace_level_server=16 trace_directory_server=/scratch/sqlnet/

7.

Use the following command to verify that you can connect through SQL*Plus:
>sqlplus 'aso/1234@(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=oracleserver.mydomain.com)( PORT=5529)) (CONNECT_DATA=(SERVICE_NAME=mydatabaseinstance)))'

Code Example
This example demonstrates the new RADIUS authentication feature that is a part of Oracle Database 11g Release 1 (11.1) JDBC thin driver. You need to have a working setup, that is, a RADIUS server up and running, and an Oracle database server that is configured to use RADIUS authentication. You then need to change the URLs given in the example to compile and run it.
Example 95 import import import import import java.sql.Connection; java.sql.ResultSet; java.sql.SQLException; java.sql.Statement; java.util.Properties;

JDBC Client-Side Security Features 9-19

Support for RADIUS

import oracle.jdbc.OracleConnection; import oracle.jdbc.OracleDriver; import oracle.net.ano.AnoServices; public class RadiusJdbcDemo { String url ="jdbc:oracle:thin:@(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)"+ "(HOST=oracleserver.mydomain.com)(PORT=5561))(CONNECT_DATA=" + "(SERVICE_NAME=mydatabaseinstance)))"; public static void main(String[] arv) { RadiusJdbcDemo radiusDemo = new RadiusJdbcDemo(); try { radiusDemo.connect(); } catch (Exception e) { e.printStackTrace(); } } /* * This method attempts to logon to the database using the RADIUS * authentication protocol. * * It should print the following output to stdout: * ----------------------------------------------------* Authentication adaptor=RADIUS * User is:ASO * ----------------------------------------------------*/ void connect() throws SQLException { OracleDriver driver = new OracleDriver(); Properties prop = new Properties();

prop.setProperty(OracleConnection.CONNECTION_PROPERTY_THIN_NET_AUTHENTICATION_SERV ICES, "("+AnoServices.AUTHENTICATION_RADIUS+")"); // The user "aso" needs to be properly setup on the radius server with // password "1234". prop.setProperty("user","aso"); prop.setProperty("password","1234"); Connection conn = driver.connect(url,prop); String auth = ((OracleConnection)conn).getAuthenticationAdaptorName(); System.out.println("Authentication adaptor="+auth); printUserName(conn); conn.close(); }

void printUserName(Connection conn) throws SQLException { Statement stmt = null; try { stmt = conn.createStatement();

9-20 Oracle Database JDBC Developer's Guide and Reference

Secure External Password Store

ResultSet rs = stmt.executeQuery("select user from dual"); while(rs.next()) System.out.println("User is:"+rs.getString(1)); rs.close(); } finally { if(stmt != null) stmt.close(); } } }

Secure External Password Store

As an alternative for large-scale deployments where applications use password credentials to connect to databases, it is possible to store such credentials in a client-side Oracle wallet. An Oracle wallet is a secure software container that is used to store authentication and signing credentials. Storing database password credentials in a client-side Oracle wallet eliminates the need to embed user names and passwords in application code, batch jobs, or scripts. This reduces the risk of exposing passwords in the clear in scripts and application code, and simplifies maintenance because you need not change your code each time user names and passwords change. In addition, not having to change application code also makes it easier to enforce password management policies for these user accounts. When you configure a client to use the external password store, applications can use the following syntax to connect to databases that use password authentication:
CONNECT /@database_alias

Note that you need not specify database login credentials in this CONNECT statement. Instead your system looks for database login credentials in the client wallet.
See Also: Oracle Database Advanced Security Administrator's Guide for information about configuring your client to use secure external password store and for information about managing credentials in it.

JDBC Client-Side Security Features 9-21

Secure External Password Store

9-22 Oracle Database JDBC Developer's Guide and Reference

10
Proxy Authentication
Oracle Java Database Connectivity (JDBC) provides proxy authentication, also called N-tier authentication. This feature is supported through both the JDBC Oracle Call Interface (OCI) driver and the JDBC Thin driver. This chapter contains the following sections:

About Proxy Authentication Types of Proxy Connections Creating Proxy Connections Closing a Proxy Session Caching Proxy Connections
Note:

Oracle Database supports proxy authentication functionality in three tiers only. It does not support it across multiple middle tiers.

About Proxy Authentication

Proxy authentication is the process of using a middle-tier for user authentication. You can design a middle-tier server to proxy clients in a secure fashion by using the following three forms of proxy authentication:

The middle-tier server authenticates itself with the database server and a client. In this case, an application user or another application, authenticates itself with the middle-tier server. Client identities can be maintained all the way through to the database. The client, that is, a database user, is not authenticated by the middle-tier server. The client's identity and database password are passed through the middle-tier server to the database server for authentication. The client, that is, a global user, is authenticated by the middle-tier server, and passes either a Distinguished name (DN) or a Certificate through the middle tier for retrieving the client's user name.
See Also:

"Creating Proxy Connections" on page 10-3

In all cases, an administrator must authorize the middle-tier server to proxy a client, that is, to act on behalf of the client. Operations done on behalf of a client by a middle-tier server can be audited. Issue the following command to authorize the middle-tier server to proxy a client:
ALTER USER jeff GRANT CONNECT THROUGH scott;

Proxy Authentication 10-1

Types of Proxy Connections

where, scott is the name of the proxy user. You can also:

Specify roles that the middle tier is permitted to activate when connecting as the client. For example,
CREATE ROLE role1; GRANT SELECT ON emp TO role1;

The role clause limits the access only to those database objects that are mentioned in the list of the roles. The list of roles can be empty.

Find the users who are currently authorized to connect through a middle tier by querying the PROXY_USERS data dictionary view. Disallow a proxy connection by using the REVOKE CONNECT THROUGH clause of ALTER USER command.

Note:

In this chapter, a JDBC connection to a database is a user session in the database and vice versa.

You need to use the different fields and methods present in the oracle.jdbc.OracleConnection interface to set up the different types of proxy connections.

Types of Proxy Connections

You can create proxy connections using any one of the following options:

USER NAME This is done by supplying the user name or the password or both. The SQL statement for specifying authentication using password is:
ALTER USER jeff GRANT CONNECT THROUGH scott AUTHENTICATED USING password;

In this case, jeff is the user name and scott is the proxy for jeff. The password option exists for additional security. Having no authenticated clause implies default authentication, which is using only the user name without the password. The SQL statement for specifying default authentication is:
ALTER USER jeff GRANT CONNECT THROUGH scott

DISTINGUISHED NAME This is a global name in lieu of the password of the user being proxied for. An example of the corresponding SQL statement using a distinguished name is:
CREATE USER jeff IDENTIFIED GLOBALLY AS 'CN=jeff,OU=americas,O=oracle,L=redwoodshores,ST=ca,C=us';

The string that follows the identified globally as clause is the distinguished name. It is then necessary to authenticate using this distinguished name. The corresponding SQL statement to specify authentication using distinguished name is:
ALTER USER jeff GRANT CONNECT THROUGH scott AUTHENTICATED USING DISTINGUISHED NAME; 10-2 Oracle Database JDBC Developer's Guide and Reference

Creating Proxy Connections

CERTIFICATE This is a more encrypted way of passing the credentials of the user, who is to be proxied, to the database. The certificate contains the distinguished name encoded in it. One way of generating the certificate is by creating a wallet and then decoding the wallet to get the certificate. The wallet can be created using runutl mkwallet. It is then necessary to authenticate using the generated certificate. The SQL statement for specifying authentication using certificate is:
ALTER USER jeff GRANT CONNECT THROUGH scott AUTHENTICATED USING CERTIFICATE; .

Note:

The use of certificates for proxy authentication will be desupported in future Oracle Database releases.

Note:

All the options can be associated with roles. When opening a new proxied connection, a new session is started on the database server. Along with this session a new local transaction is created.

Creating Proxy Connections

A user, say jeff, has to connect to the database through another user, say scott. The proxy user, scott, should have an active authenticated connection. A proxy session is then created on this active connection, with the driver issuing a command to the server to create a session for the user, jeff. The server returns the new session id, and the driver sends a session switch command to switch to this new session. The JDBC OCI and Thin driver switch sessions in the same manner. The drivers permanently switch to the new session, jeff. As a result, the proxy session, scott, is not available until the new session, jeff, is closed.
Note:

You can use the isProxySession method from the oracle.jdbc.OracleConnection interface to check if the current session associated with your connection is a proxy session. This method returns true if the current session associated with the connection is a proxy session.

A new proxy session is opened by using the following method from the oracle.jdbc.OracleConnection interface:
void openProxySession(int type, java.util.Properties prop) throws SQLExceptionOpens

Where, type is the type of the proxy session and can have the following values:

OracleConnection.PROXYTYPE_USER_NAME This type is used for specifying the user name.

OracleConnection.PROXYTYPE_DISTINGUISHED_NAME
Proxy Authentication 10-3

Closing a Proxy Session

This type is used for specifying the distinguished name of the user.

OracleConnection.PROXYTYPE_CERTIFICATE This type is used for specifying the proxy certificate.

prop is the property value of the proxy session and can have the following values:

PROXY_USER_NAME This property value should be used with the type OracleConnection.PROXYTYPE_USER_NAME. The value should be a java.lang.String.

PROXY_DISTINGUISHED_NAME This property value should be used with the type OracleConnection.PROXYTYPE_DISTINGUISHED_NAME. The value should be a java.lang.String.

PROXY_CERTIFICATE This property value should be used with the type OracleConnection.PROXYTYPE_CERTIFICATE. The value is a bytep[] array that contains the certificate.

PROXY_ROLES This property value can be used with the following types: OracleConnection.PROXYTYPE_USER_NAME OracleConnection.PROXYTYPE_DISTINGUISHED_NAME OracleConnection.PROXYTYPE_CERTIFICATE

The value should be a java.lang.String.

PROXY_SESSION This property value is used with the close method to close the proxy session.
See Also:

Closing a Proxy Session on page 10-4

PROXY_USER_PASSWORD This property value should be used with the type OracleConnection.PROXYTYPE_USER_NAME. The value should be a java.lang.String.

The following code snippet shows the use of the openProxySession method:
java.util.Properties prop = new java.util.Properties(); prop.put(OracleConnection.PROXY_USER_NAME, "jeff"); String[] roles = {"role1", "role2"}; prop.put(OracleConnection.PROXY_ROLES, roles); conn.openProxySession(OracleConnection.PROXYTYPE_USER_NAME, prop);

Closing a Proxy Session

You can close the proxy session opened with the OracleConnection.openProxySession method by passing the OracleConnection.PROXY_SESSION parameter to the OracleConnection.close method in the following way:

10-4 Oracle Database JDBC Developer's Guide and Reference

Caching Proxy Connections

OracleConnection.close(OracleConnection.PROXY_SESSION);

This is similar to closing a proxy session on a non-cached connection. The standard close method must be called explicitly to close the connection itself. If the close method is called directly, without closing the proxy session, then both the proxy session and the connection are closed. This can be achieved in the following way:
OracleConnection.close(OracleConnection.INVALID_CONNECTION);

See Also:

Chapter 21, "Implicit Connection Caching"

Caching Proxy Connections

Proxy connections, like standard connections, can be cached. Caching proxy connections enhances the performance. To cache a proxy connection, you need to create a connection using one of the getConnection methods on a cache enabled OracleDataSource object.
See Also:

Chapter 21, "Implicit Connection Caching"

A proxy connection may be cached in the connection cache using the connection attributes feature of the connection cache. Connection attributes are name/value pairs that are user-defined and help tag a connection before returning it to the connection cache for reuse. When the tagged connection is retrieved, it can be directly used without having to do a round-trip to create or close a proxy session. Implicit connection cache supports caching of any user/password authenticated connection. Therefore, any user authenticated proxy connection can be cached and retrieved. It is recommended that proxy connections should not be closed without applying the connection attributes. If a proxy connection is closed without applying the connection attributes, the connection is returned to the connection cache for reuse, but cannot be retrieved. The connection caching mechanism does not remember or reset session state. A proxy connection can be removed from the connection cache by closing the connection directly.

See Also:

"Closing a Proxy Session" on page 10-4

Proxy Authentication 10-5

Caching Proxy Connections

10-6 Oracle Database JDBC Developer's Guide and Reference

Part IV
Data Access and Manipulation
This part provides a chapter that discusses about accessing and manipulating Oracle data. It also includes chapters that provide information about Java Database Connectivity (JDBC) support for user-defined object types, large object (LOB) and binary file (BFILE) locators and data, object references, and Oracle collections, such as nested tables. This part also provides chapters that discuss the result set functionality in JDBC, JDBC row sets, and globalization support provided by Oracle JDBC drivers. Part IV contains the following chapters:

Chapter 11, "Accessing and Manipulating Oracle Data" Chapter 12, "Java Streams in JDBC" Chapter 13, "Working with Oracle Object Types" Chapter 14, "Working with LOBs and BFILEs" Chapter 15, "Using Oracle Object References" Chapter 16, "Working with Oracle Collections" Chapter 17, "Result Set" Chapter 18, "JDBC RowSets" Chapter 19, "Globalization Support"

11
Accessing and Manipulating Oracle Data
This chapter describes data access in oracle.sql.* formats, as opposed to standard Java formats. Using oracle.sql.* formats involves casting your result sets and statements to OracleResultSet, OracleStatement, OraclePreparedStatement, and OracleCallableStatement, as appropriate, and using the getOracleObject, setOracleObject, getXXX, and setXXX methods of these classes, where XXX corresponds to the types in the oracle.sql package. This chapter covers the following topics:

Data Type Mappings Data Conversion Considerations Result Set and Statement Extensions Comparison of Oracle get and set Methods to Standard JDBC Using Result Set Metadata Extensions Using SQL CALL and CALL INTO Statements

Data Type Mappings

The Oracle JDBC drivers support standard JDBC types as well as Oracle-specific data types. This section documents standard and Oracle-specific SQL-Java default type mappings. This section contains the following topics:

Table of Mappings Notes Regarding Mappings

Table of Mappings
Table 111 shows the default mappings between SQL data types, JDBC type codes, standard Java types, and Oracle extended types. The SQL Data Types column lists the SQL types that exist in Oracle Database 10g. The JDBC Type Codes column lists data type codes supported by the JDBC standard and defined in the java.sql.Types class or by Oracle in the oracle.jdbc.OracleTypes class. For standard type codes, the codes are identical in these two classes. The Standard Java Types column lists standard types defined in the Java language. The Oracle Extension Java Types column lists the oracle.sql.* Java types that correspond to each SQL data type in the database. These are Oracle extensions that let you retrieve all SQL data in the form of a oracle.sql.* Java type. Manipulating
Accessing and Manipulating Oracle Data 11-1

Data Type Mappings

SQL data as oracle.sql.* data types minimizes conversions, improving performance and eliminating conversion losses.
See Also:

"Package oracle.sql" on page 4-5

Table 111

Default Mappings Between SQL Types and Java Types JDBC Type Codes STANDARD JDBC TYPES: Standard Java Types Oracle Extension Java Types

SQL Data Types

CHAR VARCHAR2 LONG NUMBER NUMBER NUMBER NUMBER NUMBER NUMBER NUMBER NUMBER NUMBER NUMBER RAW RAW LONGRAW DATE DATE TIMESTAMP BLOB CLOB user-defined object user-defined reference user-defined collection ROWID NCLOB NCHAR

java.sql.Types.CHAR java.sql.Types.VARCHAR java.sql.Types.LONGVARCHAR java.sql.Types.NUMERIC java.sql.Types.DECIMAL java.sql.Types.BIT java.sql.Types.TINYINT java.sql.Types.SMALLINT java.sql.Types.INTEGER java.sql.Types.BIGINT java.sql.Types.REAL java.sql.Types.FLOAT java.sql.Types.DOUBLE java.sql.Types.BINARY java.sql.Types.VARBINARY

java.lang.String java.lang.String java.lang.String

oracle.sql.CHAR oracle.sql.CHAR oracle.sql.CHAR

java.math.BigDecima oracle.sql.NUMBER l java.math.BigDecima oracle.sql.NUMBER l boolean byte short int long float double double byte[] byte[] oracle.sql.NUMBER oracle.sql.NUMBER oracle.sql.NUMBER oracle.sql.NUMBER oracle.sql.NUMBER oracle.sql.NUMBER oracle.sql.NUMBER oracle.sql.NUMBER oracle.sql.RAW oracle.sql.RAW oracle.sql.RAW oracle.sql.DATE oracle.sql.DATE

java.sql.Types.LONGVARBINARY byte[] java.sql.Types.DATE java.sql.Types.TIME java.sql.Types.TIMESTAMP java.sql.Types.BLOB java.sql.Types.CLOB java.sql.Types.STRUCT java.sql.Types.REF java.sql.Types.ARRAY java.sql.Types.ROWID java.sql.Types.NCLOB java.sql.Types.NCHAR ORACLE EXTENSIONS: java.sql.Date java.sql.Time

javal.sql.Timestamp oracle.sql.TIMESTAMP java.sql.Blob java.sql.Clob java.sql.Struct java.sql.Ref java.sql.Array java.sql.RowId java.sql.NClob java.lang.String oracle.sql.BLOB oracle.sql.CLOB oracle.sql.STRUCT oracle.sql.REF oracle.sql.ARRAY oracle.sql.ROWID oracle.sql.NCLOB oracle.sql.CHAR

11-2 Oracle Database JDBC Developer's Guide and Reference

Data Conversion Considerations

Table 111 (Cont.) Default Mappings Between SQL Types and Java Types SQL Data Types BFILE REF CURSOR TIMESTAMP TIMESTAMP WITH TIME ZONE TIMESTAMP WITH LOCAL TIME ZONE JDBC Type Codes Standard Java Types Oracle Extension Java Types oracle.sql.BFILE oracle.jdbc.OracleRe sultSet oracle.sql.TIMESTAMP oracle.sql.TIMESTAMP TZ oracle.sql.TIMESTAMP LTZ

oracle.jdbc.OracleTypes.BFIL NA E oracle.jdbc.OracleTypes.CURS java.sql.ResultSet OR oracle.jdbc.OracleTypes.TIME java.sql.Timestamp STAMP oracle.jdbc.OracleTypes.TIME java.sql.Timestamp STAMPTZ oracle.jdbc.OracleTypes.TIME java.sql.Timestamp STAMPLTZ

Note:

For database versions, such as 8.1.7, which do not support the TIMESTAMP data type, TIMESTAMP is mapped to DATE.

See Also :

"Valid SQL-JDBC Data Type Mappings" on page A-1 Chapter 4, "Oracle Extensions"

Notes Regarding Mappings

This section provides further detail regarding mappings for NUMBER and user-defined types. NUMBER Types For the different type codes that an Oracle NUMBER value can correspond to, call the getter routine that is appropriate for the size of the data for mapping to work properly. For example, call getByte to get a Java tinyint value for an item x, where -128 < x < 128. User-Defined Types User-defined types, such as objects, object references, and collections, map by default to weak Java types, such as java.sql.Struct, but alternatively can map to strongly typed custom Java classes. Custom Java classes can implement one of two interfaces:

The standard java.sql.SQLData The Oracle-specific oracle.sql.ORAData

See Also:

"Mapping Oracle Objects" on page 13-1 and "Creating and Using Custom Object Classes for Oracle Objects" on page 13-6

Data Conversion Considerations

When JDBC programs retrieve SQL data into Java, you can use standard Java types, or you can use types of the oracle.sql package. This section covers the following topics:

Accessing and Manipulating Oracle Data 11-3

Data Conversion Considerations

Standard Types Versus Oracle Types Converting SQL NULL Data Testing for NULLs

Standard Types Versus Oracle Types

The Oracle data types in oracle.sql store data in the same bit format as used by the database. In versions of the Oracle JDBC drivers prior to Oracle Database 10g, the Oracle data types were generally more efficient. The Oracle Database 10g JDBC drivers were substantially updated. As a result, in most cases the standard Java types are preferred to the data types in oracle.sql. In particular, java.lang.String is much more efficient than oracle.sql.CHAR. In general, Oracle recommends that you use the Java standard types. The exceptions to this are:

Use the oracle.sql.OraData rather than the java.sql.SqlData if the OraData functionality better suits your needs. Use oracle.sql.NUMBER rather than java.lang.Double if you need to retain the exact values of floating point numbers. Oracle NUMBER is a decimal representation and Java Double and Float are binary representations. Conversion from one format to the other can result in slight variations in the actual value represented. Additionally, the range of values that can be represented using the two formats is different. Use oracle.sql.DATE or oracle.sql.TIMESTAMP rather than java.sql.Date or java.sql.Timestamp if you are using JDK 1.5 or earlier versions or require maximum performance. You may also use the oracle.sql data type if you want to read many date values or compute or display only a small percentage. Due to a bug in all versions of Java prior to JDK 1.6, construction of java.lang.Date and java.lang.Timestamp objects is slow, especially in multithreaded environments. This bug is fixed in JDK 1.6.
Note:

If you convert an oracle.sql data type to a Java standard data type, then the benefits of using the oracle.sql data type are lost.

Converting SQL NULL Data

Java represents a SQL NULL datum by the Java value null. Java data types fall into two categories: primitive types, such as byte, int, and float, and object types, such as class instances. The primitive types cannot represent null. Instead, they store null as the value zero, as defined by the JDBC specification. This can lead to ambiguity when you try to interpret your results. In contrast, Java object types can represent null. The Java language defines an object container type corresponding to every primitive type that can represent null. The object container types must be used as the targets for SQL data to detect SQL NULL without ambiguity.

Testing for NULLs

You cannot use a relational operator to compare NULL values with each other or with other values. For example, the following SELECT statement does not return any row even if the COMM column contains one or more NULL values.

11-4 Oracle Database JDBC Developer's Guide and Reference

Comparison of Oracle get and set Methods to Standard JDBC

PreparedStatement pstmt = conn.prepareStatement( "SELECT * FROM EMP WHERE COMM = ?"); pstmt.setNull(1, java.sql.Types.VARCHAR);

The next example shows how to compare values for equality when some return values might be NULL. The following code returns all the ENAMES from the EMP table that are NULL, if there is no value of 100 for COMM.
PreparedStatement pstmt = conn.prepareStatement("SELECT ENAME FROM EMP WHERE COMM =? OR ((COMM IS NULL) AND (? IS NULL))"); pstmt.setBigDecimal(1, new BigDecimal(100)); pstmt.setNull(2, java.sql.Types.VARCHAR);

Result Set and Statement Extensions

The JDBC Statement object returns an OracleResultSet object, typed as a java.sql.ResultSet. If you want to apply only standard JDBC methods to the object, then keep it as a ResultSet type. However, if you want to use the Oracle extensions on the object, then you must cast it to OracleResultSet. All of the Oracle ResultSet extensions are in the oracle.jdbc.OracleResultSet interface and all the Statement extensions are in the oracle.jdbc.OracleStatement interface. For example, assuming you have a standard Statement object stmt, do the following if you want to use only standard JDBC ResultSet methods:
ResultSet rs = stmt.executeQuery("SELECT * FROM emp");

If you need the extended functionality provided by the Oracle extensions to JDBC, you can select the results into a standard ResultSet variable and then cast that variable to OracleResultSet later. Key extensions to the result set and statement classes include the getOracleObject and setOracleObject methods, used to access and manipulate data in oracle.sql.* formats.

Comparison of Oracle get and set Methods to Standard JDBC

This section describes get and set methods, particularly the JDBC standard getObject and setObject methods and the Oracle-specific getOracleObject and setOracleObject methods, and how to access data in oracle.sql.* format compared with Java format. Although there are specific getXXX methods for all the Oracle SQL types, you can use the general get methods for convenience or simplicity, or if you are not certain in advance what type of data you will receive. This section covers the following topics:

Standard getObject Method Oracle getOracleObject Method Summary of getObject and getOracleObject Return Types Other getXXX Methods Data Types For Returned Objects from getObject and getXXX The setObject and setOracleObject Methods Other setXXX Methods
Accessing and Manipulating Oracle Data 11-5

Comparison of Oracle get and set Methods to Standard JDBC

Note: You cannot qualify a column name with a table name and pass it as a parameter to the getXXX method. For example:
ResultSet rset = stmt.executeQuery("SELECT emp.deptno, dept.deptno FROM emp, dept"); rset.getInt("emp.deptno");

The getInt method in the preceding code will throw an exception. To uniquely identify the columns in the getXXX method, you can either use column index or specify column aliases in the query and use these aliases in the getXXX method.

Standard getObject Method

The standard getObject method of a result set or callable statement has a return type of java.lang.Object. The class of the object returned is based on its SQL type, as follows:

For SQL data types that are not Oracle-specific, getObject returns the default Java type corresponding to the SQL type of the column, following the mapping in the JDBC specification. For Oracle-specific data types, getObject returns an object of the appropriate oracle.sql.* class, such as oracle.sql.ROWID. For Oracle database objects, getObject returns a Java object of the class specified in your type map. Type maps specify a mapping from database named types to Java classes. The getObject(parameter_index) method uses the default type map of the connection. The getObject(parameter_index, map) enables you to pass in a type map. If the type map does not provide a mapping for a particular Oracle object, then getObject returns an oracle.sql.STRUCT object.

Oracle getOracleObject Method

If you want to retrieve data from a result set or callable statement as an oracle.sql.* object, then you must follow a special process. For a ResultSet, you must cast the result set itself to oracle.jdbc.OracleResultSet and then call getOracleObject instead of getObject. The same applies to CallableStatement and oracle.jdbc.OracleCallableStatement. The return type of getOracleObject is oracle.sql.Datum. The actual returned object is an instance of the appropriate oracle.sql.* class. The method signature is:
public oracle.sql.Datum getOracleObject(int parameter_index)

When you retrieve data into a Datum variable, you can use the standard Java instanceof operator to determine which oracle.sql.* type it really is. Example: Using getOracleObject with a ResultSet The following example creates a table that contains a column of CHAR data and a column containing a BFILE locator. A SELECT statement retrieves the contents of the table as a result set. The getOracleObject then retrieves the CHAR data into the char_datum variable and the BFILE locator into the bfile_datum variable. Note that because getOracleObject returns a Datum object, the return values must be cast to CHAR and BFILE, respectively.
stmt.execute ("CREATE TABLE bfile_table (x VARCHAR2 (30), b BFILE)"); stmt.execute

11-6 Oracle Database JDBC Developer's Guide and Reference

Comparison of Oracle get and set Methods to Standard JDBC

("INSERT INTO bfile_table VALUES ('one', BFILENAME ('TEST_DIR', 'file1'))"); ResultSet rset = stmt.executeQuery ("SELECT * FROM bfile_table"); while (rset.next ()) { CHAR char_datum = (CHAR) ((OracleResultSet)rset).getOracleObject (1); BFILE bfile_datum = (BFILE) ((OracleResultSet)rset).getOracleObject (2); ... }

Example: Using getOracleObject in a Callable Statement The following example prepares a call to the procedure myGetDate, which associates a character string with a date. The program passes "SCOTT" to the prepared call and registers the DATE type as an output parameter. After the call is run, getOracleObject retrieves the date associated with "SCOTT". Note that because getOracleObject returns a Datum object, the results are cast to DATE.
OracleCallableStatement cstmt = (OracleCallableStatement)conn.prepareCall ("begin myGetDate (?, ?); end;"); cstmt.setString (1, "SCOTT"); cstmt.registerOutParameter (2, Types.DATE); cstmt.execute (); DATE date = (DATE) ((OracleCallableStatement)cstmt).getOracleObject (2); ...

Summary of getObject and getOracleObject Return Types

Table 112 lists the underlying return types for the getObject and getOracleObject methods for each Oracle SQL type. Keep in mind the following when you use these methods:

getObjectalways returns data into a java.lang.Object instance getOracleObject always returns data into an oracle.sql.Datum instance

You must cast the returned object to use any special functionality.

Table 112

getObject and getOracleObject Return Types getObject Underlying Return Type String String String String java.math.BigDecimal byte[] byte[] java.sql.Date java.sql.Timestamp
1

Oracle SQL Type CHAR VARCHAR2 NCHAR LONG NUMBER RAW LONGRAW DATE TIMESTAMP

getOracleObject Underlying Return Type oracle.sql.CHAR oracle.sql.CHAR oracle.sql.CHAR oracle.sql.CHAR oracle.sql.NUMBER oracle.sql.RAW oracle.sql.RAW oracle.sql.DATE oracle.sql.TIMESTAMP

Accessing and Manipulating Oracle Data 11-7

Comparison of Oracle get and set Methods to Standard JDBC

Table 112 (Cont.) getObject and getOracleObject Return Types Oracle SQL Type TIMESTAMP WITH TIME ZONE TIMESTAMP WITH LOCAL TIME ZONE BINARY_FLOAT BINARY_DOUBLE INTERVAL DAY TO SECOND INTERVAL YEAR TO MONTH ROWID REF CURSOR BLOB CLOB NCLOB BFILE Oracle object getObject Underlying Return Type oracle.sql.TIMESTAMPTZ oracle.sql.TIMESTAMPLTZ java.lang.Float java.lang.Double oracle.sql.INTERVALDS oracle.sql.INTERVALYM oracle.sql.ROWID java.sql.ResultSet oracle.sql.BLOB oracle.sql.CLOB java.sql.NClob oracle.sql.BFILE class specified in type map or oracle.sql.STRUCT (if no type map entry) Oracle object reference collection (varray or nested table)
1

getOracleObject Underlying Return Type oracle.sql.TIMESTAMPTZ oracle.sql.TIMESTAMPLTZ oracle.sql.BINARY_FLOAT oracle.sql.BINARY_DOUBLE oracle.sql.INTERVALDS oracle.sql.INTERVALYM oracle.sql.ROWID (not supported) oracle.sql.BLOB oracle.sql.CLOB oracle.sql.NCLOB oracle.sql.BFILE oracle.sql.STRUCT

oracle.sql.REF oracle.sql.ARRAY

oracle.sql.REF oracle.sql.ARRAY

ResultSet.getObject returns java.sql.Timestamp only if the oracle.jdbc.J2EE13Compliant connection property is set to TRUE, else the method returns oracle.sql.TIMESTAMP.

11-8 Oracle Database JDBC Developer's Guide and Reference

Comparison of Oracle get and set Methods to Standard JDBC

The ResultSet.getObject method returns java.sql.Timestamp for the TIMESTAMP SQL type, only when the connection property oracle.jdbc.J2EE13Compliant is set to TRUE. This property has to be set when the connection is obtained. If this connection property is not set or if it is set after the connection is obtained, then the ResultSet.getObject method returns oracle.sql.TIMESTAMP for the TIMESTAMP SQL type.
Note:

The oracle.jdbc.J2EE13Compliant connection property can also be set without changing the code in the following ways:

Including the ojdbc5dms.jar or ojdbc6dms.jar files in the CLASSPATH. These files set oracle.jdbc.J2EE13Compliant to TRUE by default. These are specific to the Oracle Application Server release and are not available as part of the general JDBC release. They are located in $ORACLE_HOME/jdbc/lib. Setting the system property by calling the java command with the flag -Doracle.jdbc.J2EE13Compliant=true. For example,
java -Doracle.jdbc.J2EE13Compliant=true ...

When the J2EE13Compliant is set to TRUE the action is as in Table B-3 of the JDBC specification.
See Also: Table A1, " Valid SQL Data Type-Java Class Mappings" on page A-1, for information about type compatibility between all SQL and Java types.

Other getXXX Methods

Standard JDBC provides a getXXX for each standard Java type, such as getByte, getInt, getFloat, and so on. Each of these returns exactly what the method name implies. In addition, the OracleResultSet and OracleCallableStatement classes provide a full complement of getXXX methods corresponding to all the oracle.sql.* types. Each getXXX method returns an oracle.sql.XXX object. For example, getROWID returns an oracle.sql.ROWID object. There is no performance advantage in using the specific getXXX methods. However, they do save you the trouble of casting, because the return type is specific to the object being returned. This section covers the following topics:

Return Types of getXXX Methods Special Notes about getXXX Methods

Return Types of getXXX Methods

Refer to the Java doc to know the return types for each getXXX method and also which are Oracle extensions under Java Development Kit (JDK) 1.6. You must cast the returned object to OracleResultSet or OracleCallableStatement to use methods that are Oracle extensions.

Accessing and Manipulating Oracle Data 11-9

Comparison of Oracle get and set Methods to Standard JDBC

Special Notes about getXXX Methods

This section provides additional details about some getXXX methods. getBigDecimal JDBC 2.0 simplified method signatures for the getBigDecimal method. The previous input signatures were:
(int columnIndex, int scale) or (String columnName, int scale)

The simplified input signature is:

(int columnIndex) or (String columnName)

The scale parameter, used to specify the number of digits to the right of the decimal, is no longer necessary. The Oracle JDBC drivers retrieve numeric values with full precision. getBoolean Because there is no BOOLEAN database type, when you use getBoolean a data type conversion always occurs. The getBoolean method is supported only for numeric columns. When applied to these columns, getBoolean interprets any zero value as false and any other value as true. When applied to any other sort of column, getBoolean raises the exception java.lang.NumberFormatException.

Data Types For Returned Objects from getObject and getXXX

The return type of getObject is java.lang.Object. The returned value is an instance of a subclass of java.lang.Object. Similarly, the return type of getOracleObject is oracle.sql.Datum, and the class of the returned value is a subclass of oracle.sql.Datum. You typically cast the returned object to the appropriate class to use particular methods and functionality of that class. In addition, you have the option of using a specific getXXX method instead of the generic getObject or getOracleObject methods. The getXXX methods enable you to avoid casting, because the return type of getXXX corresponds to the type of object returned. For example, the return type of getCLOB is oracle.sql.CLOB, as opposed to java.lang.Object. Example of Casting Return Values This example assumes that you have fetched data of the NUMBER type as the first column of a result set. Because you want to manipulate the NUMBER data without losing precision, cast your result set to OracleResultSet and use getOracleObject to return the NUMBER data in oracle.sql.* format. If you do not cast your result set, then you have to use getObject, which returns your numeric data into a Java Float and loses some of the precision of your SQL data. The getOracleObject method returns an oracle.sql.NUMBER object into an oracle.sql.Datum return variable unless you cast the output. Cast the getOracleObject output to oracle.sql.NUMBER if you want to use a NUMBER return variable and any of the special functionality of that class.
NUMBER x = (NUMBER)ors.getOracleObject(1);

11-10 Oracle Database JDBC Developer's Guide and Reference

Comparison of Oracle get and set Methods to Standard JDBC

The setObject and setOracleObject Methods

Just as there is a standard getObject and Oracle-specific getOracleObject in result sets and callable statements, there are also standard setObject and Oracle-specific setOracleObject methods in OraclePreparedStatement and OracleCallableStatement. The setOracleObject methods take oracle.sql.* input parameters. To bind standard Java types to a prepared statement or callable statement, use the setObject method, which takes a java.lang.Object as input. The setObject method does support a few of the oracle.sql.* types. However, the method has been implemented so that you can enter instances of the oracle.sql.* classes that correspond to the following JDBC standard types: Blob, Clob, Struct, Ref, and Array. To bind oracle.sql.* types to a prepared statement or callable statement, use the setOracleObject method, which takes a subclass of oracle.sql.Datum as input. To use setOracleObject, you must cast your prepared statement or callable statement to OraclePreparedStatement or OracleCallableStatement. Example of Using setObject and setOracleObject For a prepared statement, the setOracleObject method binds the oracle.sql.CHAR data represented by the charVal variable to the prepared statement. To bind the oracle.sql.* data, the prepared statement must be cast to OraclePreparedStatement. Similarly, the setObject method binds the Java String data represented by the variable strVal.
PreparedStatement ps= conn.prepareStatement("text_of_prepared_statement"); ((OraclePreparedStatement)ps).setOracleObject(1,charVal); ps.setObject(2,strVal);

Other setXXX Methods

As with the getXXX methods, there are several specific setXXX methods. Standard setXXX methods are provided for binding standard Java types, and Oracle-specific setXXX methods are provided for binding Oracle-specific types. Similarly, there are two forms of the setNull method:

void setNull(int parameterIndex, int sqlType) This is specified in the standard java.sql.PreparedStatement interface. This signature takes a parameter index and a SQL type code defined by the java.sql.Types or oracle.jdbc.OracleTypes class. Use this signature to set an object other than a REF, ARRAY, or STRUCT to NULL.

void setNull(int parameterIndex, int sqlType, String sql_type_name) With JDBC 2.0, this signature is also specified in the standard java.sql.PreparedStatement interface. This method takes a SQL type name in addition to a parameter index and a SQL type code. Use this method when the SQL type code is java.sql.Types.REF, ARRAY, or STRUCT. If the type code is other than REF, ARRAY, or STRUCT, then the given SQL type name is ignored.

Similarly, the registerOutParameter method has a signature for use with REF, ARRAY, or STRUCT data:
void registerOutParameter (int parameterIndex, int sqlType, String sql_type_name)

Accessing and Manipulating Oracle Data 11-11

Comparison of Oracle get and set Methods to Standard JDBC

Binding Oracle-specific types using the appropriate setXXX methods, instead of the methods used for binding standard Java types, may offer some performance advantage. This section covers the following topics:

Input Data Binding Method setFixedCHAR for Binding CHAR Data into WHERE Clauses

Input Data Binding

There are three way to bind data for input:

Direct binding where the data itself is placed in a bind buffer Stream binding where the data is streamed LOB binding where a temporary lob is created, the data placed in the LOB using the LOB APIs, and the bytes of the LOB locator are placed in the bind buffer

The three kinds of binding have some differences in performance and have an impact on batching. Direct binding is fast and batching is fine. Stream binding is slower, may require multiple round trips, and turns batching off. LOB binding is very slow and requires many round trips. Batching works, but might be a bad idea. They also have different size limits, depending on the type of the SQL statement. For SQL parameters, the length of standard parameter types, such as RAW and VARCHAR2, is fixed by the size of the target column. For PL/SQL parameters, the size is limited to a fixed number of bytes, which is 32766. In Oracle Database 10g release 2 (10.2), certain changes were made to the setString, setCharacterStream, setAsciiStream, setBytes, and setBinaryStream methods of PreparedStatement. The original behavior of these APIs were:

setString: Direct bind of characters setCharacterStream: Stream bind of characters setAsciiStream: Stream bind of bytes setBytes: Direct bind of bytes setBinaryStream: Stream bind of bytes

Starting from Oracle Database 10g release 2 (10.2), automatic switching between binding modes, based on the data size and on the type of the SQL statement is provided. setBytes and setBinaryStream For SQL, direct bind is used for size up to 2000 and stream bind for larger. For PL/SQL direct bind is used for size up to 32766 and LOB bind is used for larger. setString, setCharacterStream, and setAsciiStream For SQL, direct bind is used up to 32766 Java characters and stream bind is used for larger. This is independent of character set. For PL/SQL, you must be careful about the byte size of the character data in the database character set or the national character set depending on the setting of the form of use parameter. Direct bind is used for data where the byte length is less than 32766 and LOB bind is used for larger.

11-12 Oracle Database JDBC Developer's Guide and Reference

Comparison of Oracle get and set Methods to Standard JDBC

For fixed length character sets, multiply the length of the Java character data by the fixed character size in bytes and compare that to the restrictive values. For variable length character sets, there are three cases based on the Java character length, as follows:

If character length is less than 32766 divided by the maximum character size, then direct bind is used. If character length is greater than 32766 divided by the minimum character size, then LOB bind is used. If character length is in between and if the actual length of the converted bytes is less than 32766, then direct bind is used, else LOB bind is used. When a PL/SQL procedure is embedded in a SQL statement, the binding action is different. Refer to "Data Interface for LOBs" on page 14-10 for more information.
Note:

The server-side internal driver has the following additional limitations:

setString, setCharacterStream, and setASCIIStream APIs are not supported for SQL CLOB columns when the data size in characters is over 4000 bytes setBytes and setBinaryStream APIs are not supported for SQL BLOB columns when the data size is over 2000 bytes
Important: Do not use these APIs with the server-side internal

driver, without careful checking of the datasize in client code.

See Also: JDBC Release Notes for further discussion and possible workarounds

Method setFixedCHAR for Binding CHAR Data into WHERE Clauses

CHAR data in the database is padded to the column width. This leads to a limitation in using the setCHAR method to bind character data into the WHERE clause of a SELECT statement. The character data in the WHERE clause must also be padded to the column width to produce a match in the SELECT statement. This is especially troublesome if you do not know the column width. To remedy this, Oracle has added the setFixedCHAR method to the OraclePreparedStatement class. This method runs a non-padded comparison.
Note:

Remember to cast your prepared statement object to OraclePreparedStatement to use the setFixedCHAR method. There is no need to use setFixedCHAR for an INSERT statement. The database always automatically pads the data to the column width as it inserts it.

Accessing and Manipulating Oracle Data 11-13

Using Result Set Metadata Extensions

Example The following example demonstrates the difference between the setCHAR and setFixedCHAR methods.
/* Schema is : create table my_table (col1 char(10)); insert into my_table values ('JDBC'); */ PreparedStatement pstmt = conn.prepareStatement ("select count(*) from my_table where col1 = ?"); pstmt.setString (1, "JDBC"); runQuery (pstmt); // Set the Bind Value // This will print " No of rows are 0"

CHAR ch = new CHAR("JDBC ", null); ((OraclePreparedStatement)pstmt).setCHAR(1, ch); // Pad it to 10 bytes runQuery (pstmt); // This will print "No of rows are 1" ((OraclePreparedStatement)pstmt).setFixedCHAR(1, "JDBC"); runQuery (pstmt); // This will print "No of rows are 1" void runQuery (PreparedStatement ps) { // Run the Query ResultSet rs = pstmt.executeQuery (); while (rs.next()) System.out.println("No of rows are " + rs.getInt(1)); rs.close(); rs = null; }

Using Result Set Metadata Extensions

The oracle.jdbc.OracleResultSetMetaData interface is JDBC 2.0-compliant but does not implement the getSchemaName and getTableName methods because Oracle Database does not make this feasible.. The following code snippet uses several of the methods in the OracleResultSetMetadata interface to retrieve the number of columns from the EMP table and the numerical type and SQL type name of each column:
DatabaseMetaData dbmd = conn.getMetaData(); ResultSet rset = dbmd.getTables("", "SCOTT", "EMP", null); while (rset.next()) { OracleResultSetMetaData orsmd = ((OracleResultSet)rset).getMetaData(); int numColumns = orsmd.getColumnCount(); System.out.println("Num of columns = " + numColumns); for (int i=0; i<numColumns; i++) { System.out.print ("Column Name=" + orsmd.getColumnName (i+1)); System.out.print (" Type=" + orsmd.getColumnType (i + 1) ); System.out.println (" Type Name=" + orsmd.getColumnTypeName (i + 1)); } }

11-14 Oracle Database JDBC Developer's Guide and Reference

Using SQL CALL and CALL INTO Statements

The program returns the following output:

Num of Column Column Column Column Column columns = 5 Name=TABLE_CAT Type=12 Type Name=VARCHAR2 Name=TABLE_SCHEM Type=12 Type Name=VARCHAR2 Name=TABLE_NAME Type=12 Type Name=VARCHAR2 Name=TABLE_TYPE Type=12 Type Name=VARCHAR2 Name=TABLE_REMARKS Type=12 Type Name=VARCHAR2

Using SQL CALL and CALL INTO Statements

You can use the CALL statement to execute a routine from within SQL.
Note:

A routine is a procedure or a function that is standalone or is defined within a type or package. You must have EXECUTE privilege on the standalone routine or on the type or package in which the routine is defined. Refer to the "Oracle Database SQL Language Reference" for more information about using the CALL statement.

You can execute a routine in two ways:

By issuing a call to the routine itself by name or by using the routine_clause By using an object_access_expression inside the type of an expression

You can specify one or more arguments to the routine, if the routine takes arguments. You can use positional, named, or mixed notation for argument. CALL INTO Statement The INTO clause applies only to calls to functions. You can use the following types of variables with this clause:

Host variable Indicator variable

PL/SQL Blocks The basic unit in PL/SQL is a block. All PL/SQL programs are made up of blocks, which can be nested within each other. A PL/SQL block has three parts: a declarative part, an executable part, and an exception-handling part. You get the following advantages by using PL/SQL blocks in your application:

Better performance Higher productivity Full portability Tight integration with Oracle Tight security

Accessing and Manipulating Oracle Data 11-15

Using SQL CALL and CALL INTO Statements

11-16 Oracle Database JDBC Developer's Guide and Reference

12
Java Streams in JDBC
This chapter describes how the Oracle Java Database Connectivity (JDBC) drivers handle Java streams for several data types. Data streams enable you to read LONG column data of up to 2 gigabytes (GB). Methods associated with streams let you read the data incrementally. This chapter covers the following topics:

Streaming LONG or LONG RAW Columns Streaming CHAR, VARCHAR, or RAW Columns Streaming LOBs and External Files Data Streaming and Multiple Columns Streaming and Row Prefetching Closing a Stream Notes and Precautions on Streams

Overview
Oracle JDBC drivers support the manipulation of data streams in either direction between server and client. The drivers support all stream conversions: binary, ASCII, and Unicode. Following is a brief description of each type of stream:

Binary Used for RAW bytes of data, and corresponds to the getBinaryStream method

ASCII Used for ASCII bytes in ISO-Latin-1 encoding, and corresponds to the getAsciiStream method

Unicode Used for Unicode bytes with the UTF-16 encoding, and corresponds to the getUnicodeStream method

The getBinaryStream, getAsciiStream, and getUnicodeStream methods return the bytes of data in an InputStream object.
See Also:

Chapter 14, "Working with LOBs and BFILEs"

Java Streams in JDBC

12-1

Streaming LONG or LONG RAW Columns

Streaming LONG or LONG RAW Columns

When a query selects one or more LONG or LONG RAW columns, the JDBC driver transfers these columns to the client in streaming mode. After a call to executeQuery or next, the data of the LONG column is waiting to be read.
Note:

Oracle recommends avoiding LONG and LONG RAW columns. Use LOB instead.

To access the data in a LONG column, you can get the column as a Java InputStream object and use the read method of the InputStream object. As an alternative, you can get the data as a String or byte array. In this case, the driver will do the streaming for you. You can get LONG and LONG RAW data with any of the three stream types. The driver performs conversions for you, depending on the character set of the database and the driver.
Note: Do not create tables with LONG columns. Use large object (LOB) columns, CLOB, NCLOB, and BLOB, instead. LONG columns are supported only for backward compatibility. Oracle recommends that you convert existing LONG columns to LOB columns. LOB columns are subject to far fewer restrictions than LONG columns.

This section covers the following topics:

Streaming LONG or LONG RAW Columns Streaming CHAR, VARCHAR, or RAW Columns Streaming LOBs and External Files Data Streaming and Multiple Columns Closing a Stream Notes and Precautions on Streams

LONG RAW Data Conversions

A call to getBinaryStream returns RAW data. A call to getAsciiStream converts the RAW data to hexadecimal and returns the ASCII representation. A call to getUnicodeStream converts the RAW data to hexadecimal and returns the Unicode characters.

LONG Data Conversions

When you get LONG data with getAsciiStream, the drivers assume that the underlying data in the database uses an US7ASCII or WE8ISO8859P1 character set. If the assumption is true, then the drivers return bytes corresponding to ASCII characters. If the database is not using an US7ASCII or WE8ISO8859P1 character set, a call to getAsciiStream returns meaningless information. When you get LONG data with getUnicodeStream, you get a stream of Unicode characters in the UTF-16 encoding. This applies to all underlying database character sets that Oracle supports. When you get LONG data with getBinaryStream, there are two possible cases:
12-2 Oracle Database JDBC Developer's Guide and Reference

Streaming LONG or LONG RAW Columns

If the driver is JDBC OCI and the client character set is not US7ASCII or WE8ISO8859P1, then a call to getBinaryStream returns UTF-8. If the client character set is US7ASCII or WE8ISO8859P1, then the call returns a US7ASCII stream of bytes. If the driver is JDBC Thin and the database character set is not US7ASCII or WE8ISO8859P1, then a call to getBinaryStream returns UTF-8. If the server-side character set is US7ASCII or WE8ISO8859P1, then the call returns a US7ASCII stream of bytes.
Tip: Chapter 19, "Globalization Support" and "Data Streaming and Multiple Columns" on page 12-7

Note: Receiving LONG or LONG RAW columns as a stream requires you to pay special attention to the order in which you retrieve columns from the database.

Table 121 summarizes LONG and LONG RAW data conversions for each stream type.
Table 121 Data type LONG LONG and LONG RAW Data Conversions BinaryStream Bytes representing characters in Unicode UTF-8. The bytes can represent characters in US7ASCII or WE8ISO8859P1 if the database character set is US7ASCII or WE8ISO8859P1. unchanged data AsciiStream Bytes representing characters in ISO-Latin-1 (WE8ISO8859P1) encoding UnicodeStream Bytes representing characters in Unicode UTF-16 encoding

LONG RAW

ASCII representation of hexadecimal bytes

Unicode representation of hexadecimal bytes

Streaming Example for LONG RAW Data

One of the features of a getXXXStream method is that it enables you to fetch data incrementally. In contrast, getBytes fetches all the data in one call. This section contains two examples of getting a stream of binary data. The first version uses the getBinaryStream method to obtain LONG RAW data, and the second version uses the getBytes method. Getting a LONG RAW Data Column with getBinaryStream This example writes the contents of a LONG RAW column to a file on the local file system. In this case, the driver fetches the data incrementally. The following code creates the table that stores a column of LONG RAW data associated with the name LESLIE:
-- SQL code: create table streamexample (NAME varchar2 (256), GIFDATA long raw); insert into streamexample values ('LESLIE', '00010203040506070809');

The following Java code snippet writes the data from the LONG RAW column into a file called leslie.gif:
ResultSet rset = stmt.executeQuery ("select GIFDATA from streamexample where NAME='LESLIE'"); // get first row

Java Streams in JDBC

12-3

Streaming LONG or LONG RAW Columns

if (rset.next()) { // Get the GIF data as a stream from Oracle to the client InputStream gif_data = rset.getBinaryStream (1); try { FileOutputStream file = null; file = new FileOutputStream ("leslie.gif"); int chunk; while ((chunk = gif_data.read()) != -1) file.write(chunk); } catch (Exception e) { String err = e.toString(); System.out.println(err); } finally { if file != null() file.close(); } }

In this example, the InputStream object returned by the call to getBinaryStream reads the data directly from the database connection. Getting a LONG RAW Data Column with getBytes This example gets the content of the GIFDATA column with getBytes instead of getBinaryStream. In this case, the driver fetches all the data in one call and stores it in a byte array. The code snippet is as follows:
ResultSet rset2 = stmt.executeQuery ("select GIFDATA from streamexample where NAME='LESLIE'"); // get first row if (rset2.next()) { // Get the GIF data as a stream from Oracle to the client byte[] bytes = rset2.getBytes(1); try { FileOutputStream file = null; file = new FileOutputStream ("leslie2.gif"); file.write(bytes); } catch (Exception e) { String err = e.toString(); System.out.println(err); } finally { if file != null() file.close(); } }

Because a LONG RAW column can contain up to 2 gigabytes of data, the getBytes example can use much more memory than the getBinaryStream example. Use
12-4 Oracle Database JDBC Developer's Guide and Reference

Streaming CHAR, VARCHAR, or RAW Columns

streams if you do not know the maximum size of the data in your LONG or LONG RAW columns.

Avoiding Streaming for LONG or LONG RAW

The JDBC driver automatically streams any LONG and LONG RAW columns. However, there may be situations where you want to avoid data streaming. For example, if you have a very small LONG column, then you may want to avoid returning the data incrementally and, instead, return the data in one call. To avoid streaming, use the defineColumnType method to redefine the type of the LONG column. For example, if you redefine the LONG or LONG RAW column as VARCHAR or VARBINARY type, then the driver will not automatically stream the data. If you redefine column types with defineColumnType, then you must declare the types of the columns in the query. If you do not declare the types of the columns, then executeQuery will fail. In addition, you must cast the Statement object to oracle.jdbc.OracleStatement. As an added benefit, using defineColumnType saves the OCI and KPRB drivers a database round-trip when running the query. Without defineColumnType, these JDBC drivers must request the data types of the column types. The JDBC Thin driver derives no benefit from defineColumnType, because it always uses the minimum number of round-trips. Using the example from the previous section, the Statement object stmt is cast to OracleStatement and the column containing LONG RAW data is redefined to be of the type VARBINARAY. The data is not streamed. Instead, it is returned in a byte array. The code snippet is as follows:
//cast the statement stmt to an OracleStatement oracle.jdbc.OracleStatement ostmt = (oracle.jdbc.OracleStatement)stmt; //redefine the LONG column at index position 1 to VARBINARY ostmt.defineColumnType(1, Types.VARBINARY); // Do a query to get the images named 'LESLIE' ResultSet rset = ostmt.executeQuery ("select GIFDATA from streamexample where NAME='LESLIE'"); // The data is not streamed here rset.next(); byte [] bytes = rset.getBytes(1);

Streaming CHAR, VARCHAR, or RAW Columns

If you use the defineColumnType Oracle extension to redefine a CHAR, VARCHAR, or RAW column as a LONGVARCHAR or LONGVARBINARY, then you can get the column as a stream. The program will behave as if the column were actually of type LONG or LONG RAW. Note that there is not much point to this, because these columns are usually short. If you try to get a CHAR, VARCHAR, or RAW column as a data stream without redefining the column type, then the JDBC driver will return a Java InputStream, but no real streaming occurs. In the case of these data types, the JDBC driver fully fetches the data into an in-memory buffer during a call to the executeQuery method or the next

Java Streams in JDBC

12-5

Streaming LOBs and External Files

method. The getXXXStream entry points return a stream that reads data from this buffer.

Streaming LOBs and External Files

The term large object (LOB) refers to a data item that is too large to be stored directly in a database table. Instead, a locator is stored in the database table, which points to the location of the actual data. External files are managed similarly. The JDBC drivers can support the following types through the use of streams:

Binary large object (BLOB) For unstructured binary data

Character large object (CLOB) For character data

National Character large object (NCLOB) For national character data

Binary file (BFILE) For external files

LOBs and BFILEs behave differently from the other types of streaming data described in this chapter. Instead of storing the actual data in the table, a locator is stored. The actual data can be manipulated using this locator, including reading and writing the data as a stream. Even when streaming, only the necessary bits of data move across the network. By contrast, when streaming a LONG or LONG RAW, all the data always moves across the network. Streaming BLOBs, CLOBs, and NCLOBs When a query fetches one or more BLOB, CLOB, or NCLOB columns, the JDBC driver transfers the data to the client. This data can be accessed as a stream. To manipulate BLOB, CLOB, or NCLOB data from JDBC, use methods in the Oracle extension classes oracle.sql.BLOB, oracle.sql.CLOB and oracle.sql.NCLOB. These classes provide specific functionality, such as reading from the BLOB, CLOB, or NCLOB into an input stream, writing from an output stream into a BLOB, CLOB, or NCLOB, determining the length of a BLOB, CLOB, or NCLOB, and closing a BLOB, CLOB, or NCLOB.
See Also: "Reading and Writing BLOB, CLOB and NCLOB Data" on page 14-5 and "Data Interface for LOBs" on page 14-10

Streaming BFILEs An external file, or BFILE, is used to store a locator to a file outside the database. The file can be stored somewhere on the file system of the data server. The locator points to the actual location of the file. When a query fetches one or more BFILE columns, the JDBC driver transfers the file to the client as required. The data can be accessed as a stream To manipulate BFILE data from JDBC, use methods in the Oracle extension class oracle.sql.BFILE. This class provides specific functionality, such as reading from the BFILE into an input stream, writing from an output stream into a BFILE, determining the length of a BFILE, and closing a BFILE.
See Also:

"Reading BFILE Data" on page 14-16

12-6 Oracle Database JDBC Developer's Guide and Reference

Data Streaming and Multiple Columns

Data Streaming and Multiple Columns

If a query fetches multiple columns and one of the columns contains a data stream, then the contents of the columns following the stream column are not available until the stream has been read, and the stream column is no longer available once any following column is read. Any attempt to read a column beyond a streaming column closes the streaming column.
See Also:

"Streaming Data Precautions" on page 12-8

Streaming Example with Multiple Columns Consider the following code:

ResultSet rset = stmt.executeQuery ("select DATECOL, LONGCOL, NUMBERCOL from TABLE"); while rset.next() { //get the date data java.sql.Date date = rset.getDate(1); // get the streaming data InputStream is = rset.getAsciiStream(2); // Open a file to store the gif data FileOutputStream file = new FileOutputStream ("ascii.dat"); // Loop, reading from the ascii stream and // write to the file int chunk; while ((chunk = is.read ()) != -1) file.write(chunk); // Close the file file.close(); //get the number column data int n = rset.getInt(3); }

The incoming data for each row has the following shape:
<a date><the characters of the long column><a number>

As you process each row of the result set, you must complete any processing of the stream column before reading the number column.
Tip:

"Streaming LOBs and External Files" on page 12-6

Bypassing Streaming Data Columns There may be situations where you want to avoid reading a column that contains streaming data. If you do not want to read such data, then call the close method of the stream object. This method discards the stream data and enables the driver to continue reading data from all the columns that contain non-streaming data and follow the column containing streaming data. Even though you are intentionally discarding the stream, it is a good programming practice to retrieve the columns in the same order as in the SELECT statement. In the following example, the stream data in the LONG column is discarded and the data from only the DATE and NUMBER column is recovered:

Java Streams in JDBC

12-7

Closing a Stream

ResultSet rset = stmt.executeQuery ("select DATECOL, LONGCOL, NUMBERCOL from TABLE"); while rset.next() { //get the date java.sql.Date date = rset.getDate(1); // access the stream data and discard it with close() InputStream is = rset.getAsciiStream(2); is.close(); // get the number column data int n = rset.getInt(3); }

Closing a Stream
You can discard the data from a stream at any time by calling the close method. It is a good programming practice to close the connection when you no longer need it.
See Also: "Bypassing Streaming Data Columns" on page 12-7 and "Streaming Data Precautions" on page 12-8

Note:

Closing a stream has little performance effect on a LONG or LONG RAW column. All of the data still move across the network and the driver can read the bits from the network.

Notes and Precautions on Streams

This section discusses several cautionary issues regarding the use of streams:

Streaming Data Precautions Using Streams to Avoid Limits on setBytes and setString Streaming and Row Prefetching

Streaming Data Precautions

This section describes some of the precautions you must take to ensure that you do not accidentally discard or lose your stream data. The drivers automatically discard stream data if you perform any JDBC operation that communicates with the database, other than reading the current stream. Two common precautions are:

Use the stream data after you access it. To recover the data from a column containing a data stream, it is not enough to fetch the column. You must immediately process the contents of the column. Otherwise, the contents will be discarded when you fetch the next column.

Call the stream column in the same order as in the SELECT statement. If your query fetches multiple columns, the database sends each row as a set of bytes representing the columns in the SELECT order. If one of the columns contains stream data, then the database sends the entire data stream before proceeding to the next column.

12-8 Oracle Database JDBC Developer's Guide and Reference

Notes and Precautions on Streams

If you do not use the order as in the SELECT statement to access data, then you can lose the stream data. That is, if you bypass the stream data column and access data in a column that follows it, then the stream data will be lost. For example, if you try to access the data for the NUMBER column before reading the data from the stream data column, then the JDBC driver first reads then discards the streaming data automatically. This can be very inefficient if the LONG column contains a large amount of data. If you try to access the LONG column later in the program, then the data will not be available and the driver will return a "Stream Closed" error. The later point is illustrated in the following example:
ResultSet rset = stmt.executeQuery ("select DATECOL, LONGCOL, NUMBERCOL from TABLE"); while rset.next() { int n = rset.getInt(3); // This discards the streaming data InputStream is = rset.getAsciiStream(2); // Raises an error: stream closed. }

If you get the stream but do not use it before you get the NUMBER column, then the stream still closes automatically:
ResultSet rset = stmt.executeQuery ("select DATECOL, LONGCOL, NUMBERCOL from TABLE"); while rset.next() { InputStream is = rset.getAsciiStream(2); // Get the stream int n = rset.getInt(3); // Discards streaming data and closes the stream } int c = is.read(); // c is -1: no more characters to read-stream closed

Using Streams to Avoid Limits on setBytes and setString

In Oracle Database 10g, the size limitation on data that may be used with setBytes and setString have been reduced and, in certain cases, eliminated. Any Java byte array can be passed to setBytes, and any Java String can be passed to setString. The JDBC driver automatically switches to using setBinaryStream or setCharacterStream or to using setBytesForBlob or setStringForClob, depending on the size of the data, whether the statement is SQL or PL/SQL, and the driver used. There are some limitation with earlier versions of Oracle Database and in the server-side internal driver.
See Also: "Data Interface for LOBs" on page 14-10 and release notes for details

Streaming and Row Prefetching

If the JDBC driver encounters a column containing a data stream, then row fetch size is set back to one. Row fetch size is an Oracle performance enhancement that allows multiple rows of data to be retrieved with each trip to the database.

Java Streams in JDBC

12-9

Notes and Precautions on Streams

12-10 Oracle Database JDBC Developer's Guide and Reference

13
Working with Oracle Object Types
This chapter describes the Java Database Connectivity (JDBC) support for user-defined object types. It discusses functionality of the generic, weakly typed oracle.sql.STRUCT class, as well as how to map to custom Java classes that implement either the JDBC standard SQLData interface or the Oracle ORAData interface. The following topics are covered:

Mapping Oracle Objects Using the Default STRUCT Class for Oracle Objects Creating and Using Custom Object Classes for Oracle Objects Object-Type Inheritance Using JPublisher to Create Custom Object Classes Describing an Object Type
See Also:

Oracle Database Object-Relational Developer's Guide

Mapping Oracle Objects

Oracle object types provide support for composite data structures in the database. For example, you can define a Person type that has the attributes name of CHAR type, phoneNumber of CHAR type, and employeeNumber of NUMBER type. Oracle provides tight integration between its Oracle object features and its JDBC functionality. You can use a standard, generic JDBC type to map to Oracle objects, or you can customize the mapping by creating custom Java type definition classes.
Note: In this book, Java classes that you create to map to Oracle objects will be referred to as custom Java classes or, more specifically, custom object classes. This is as opposed to custom references classes, which are Java classes that map to object references, and custom collection classes, which are Java classes that map to Oracle collections.

Custom object classes can implement either a standard JDBC interface or an Oracle extension interface to read and write data. JDBC materializes Oracle objects as instances of particular Java classes. Two main steps in using JDBC to access Oracle objects are:
1.

Creating the Java classes for the Oracle objects

Working with Oracle Object Types 13-1

Using the Default STRUCT Class for Oracle Objects

2.

Populating these classes. You have the following options:

Let JDBC materialize the object as a STRUCT object. Explicitly specify the mappings between Oracle objects and Java classes. This includes customizing your Java classes for object data. The driver then must be able to populate instances of the custom object classes that you specify. This imposes a set of constraints on the Java classes. To satisfy these constraints, you can define your classes to implement either the JDBC standard java.sql.SQLData interface or the Oracle extension oracle.sql.ORAData interface.

You can use the Oracle JPublisher utility to generate custom Java classes.
Note:

When you use the SQLData interface, you must use a Java type map to specify your SQL-Java mapping, unless weakly typed java.sql.Struct objects will suffice.

Using the Default STRUCT Class for Oracle Objects

If you choose not to supply a custom Java class for your SQL-Java mapping for an Oracle object, then Oracle JDBC will materialize the object as an instance of the oracle.sql.STRUCT class. You would typically want to use STRUCT objects, instead of custom Java objects, in situations where you do not know the actual SQL type. For example, your Java application might be a tool to manipulate arbitrary object data within the database, as opposed to being an end-user application. You can select data from the database into STRUCT objects and create STRUCT objects for inserting data into the database. STRUCT objects completely preserve data, because they maintain the data in SQL format. Using STRUCT objects is more efficient and more precise in situations where you do not need the information in an application specific form. This section covers the following topics:

STRUCT Class Functionality Retrieving STRUCT Objects and Attributes Creating STRUCT Objects and Descriptors Binding STRUCT Objects into Statements STRUCT Automatic Attribute Buffering

STRUCT Class Functionality

This section discusses standard versus Oracle-specific features of the oracle.sql.STRUCT class, introduces STRUCT descriptors, and lists methods of the STRUCT class to give an overview of its functionality. Standard java.sql.Struct Methods If your code must comply with standard JDBC 2.0, then use a java.sql.Struct instance and use the following standard methods:

getAttributes(map) This method retrieves the values of the attributes, using entries in the specified type map to determine the Java classes to use in materializing any attribute that is

13-2 Oracle Database JDBC Developer's Guide and Reference

Using the Default STRUCT Class for Oracle Objects

a structured object type. The Java types for other attribute values would be the same as for a getObject call on data of the underlying SQL type.

getAttributes This method is the same as the preceding getAttributes(map) method, except it uses the default type map for the connection.

getSQLTypeName This method returns a Java String that represents the fully qualified name of the Oracle object type that this Struct represents.

Oracle oracle.sql.STRUCT Class Methods If you want to take advantage of the extended functionality offered by Oracle-defined methods, then use an oracle.sql.STRUCT instance. The oracle.sql.STRUCT class implements the java.sql.Struct interface and provides extended functionality beyond the JDBC 2.0 standard. The STRUCT class includes the following methods in addition to standard Struct functionality:

getOracleAttributes Retrieves the values of the values array as oracle.sql.* objects

getDescriptor Returns the StructDescriptor object for the SQL type that corresponds to this STRUCT object

getJavaSQLConnection Returns the current connection instance

toJdbc Consults the default type map of the connection to determine what class to map to and, then, uses toClass

toJdbc(map) Consults the specified type map to determine what class to map to, and then uses toClass

Retrieving STRUCT Objects and Attributes

This section discusses how to retrieve and manipulate Oracle objects and their attributes, using either Oracle-specific features or JDBC 2.0 standard features.
Note:

The JDBC driver seamlessly handles embedded objects, that is, STRUCT objects that are attributes of STRUCT objects, in the same way that it typically handles objects. When the JDBC driver retrieves an attribute that is an object, it follows the same rules of conversion by using the type map, if it is available, or by using default mapping.

Retrieving an Oracle Object as an oracle.sql.STRUCT Object You can retrieve an Oracle object directly into an oracle.sql.STRUCT instance. In the following example, getObject is used to get a type_struct object from the

Working with Oracle Object Types 13-3

Using the Default STRUCT Class for Oracle Objects

col1 column of the table struct_table. Because getObject returns an Object type, the return is cast to oracle.sql.STRUCT. This example assumes that the Statement object stmt has already been created.
String cmd; cmd = "CREATE TYPE type_struct AS object (field1 NUMBER,field2 DATE)"; stmt.execute(cmd); cmd = "CREATE TABLE struct_table (col1 type_struct)"; stmt.execute(cmd); cmd = "INSERT INTO struct_table VALUES (type_struct(10,'01-apr-01'))"; stmt.execute(cmd); cmd = "INSERT INTO struct_table VALUES (type_struct(20,'02-may-02'))"; stmt.execute(cmd); ResultSet rs= stmt.executeQuery("SELECT * FROM struct_table"); oracle.sql.STRUCT oracleSTRUCT=(oracle.sql.STRUCT)rs.getObject(1);

Another way to return the object as a STRUCT object is to cast the result set to OracleResultSet and use the Oracle extension getSTRUCT method:
oracle.sql.STRUCT oracleSTRUCT=((OracleResultSet)rs).getSTRUCT(1);

Retrieving an Oracle Object as a java.sql.Struct Object Alternatively, in the preceding example, you can use standard JDBC functionality, such as getObject, to retrieve an Oracle object from the database as an instance of java.sql.Struct. Because getObject returns a java.lang.Object, you must cast the output of the method to Struct. For example:
ResultSet rs= stmt.executeQuery("SELECT * FROM struct_table"); java.sql.Struct jdbcStruct = (java.sql.Struct)rs.getObject(1);

Retrieving Attributes as oracle.sql Types If you want to retrieve Oracle object attributes from a STRUCT or Struct instance as oracle.sql types, then use the getOracleAttributes method of the oracle.sql.STRUCT class, as follows:
oracle.sql.Datum[] attrs = oracleSTRUCT.getOracleAttributes();

or:
oracle.sql.Datum[] attrs = ((oracle.sql.STRUCT)jdbcStruct).getOracleAttributes();

Retrieving Attributes as Standard Java Types If you want to retrieve Oracle object attributes as standard Java types from a STRUCT or Struct instance, use the standard getAttributes method:
Object[] attrs = jdbcStruct.getAttributes();

13-4 Oracle Database JDBC Developer's Guide and Reference

Using the Default STRUCT Class for Oracle Objects

Note:

Oracle JDBC drivers cache array and structure descriptors. This provides enormous performance benefits. However, it means that if you change the underlying type definition of a structure type in the database, the cached descriptor for that structure type will become stale and your application will receive a SQLException exception.

Creating STRUCT Objects and Descriptors

This section describes how to create STRUCT objects. Steps in Creating StructDescriptor and STRUCT Objects To create a STRUCT object, you must:
1. 2.

Create a StructDescriptor object for the given Oracle object type, if it does not already exist. Use the StructDescriptor to construct the STRUCT object.

A StructDescriptor is an instance of the oracle.sql.StructDescriptor class and describes a type of Oracle object. Only one StructDescriptor is necessary for each Oracle object type. The driver caches StructDescriptor objects to avoid re-creating them if the type has already been encountered. Before you can construct a STRUCT object, a StructDescriptor must first exist for the given Oracle object type. If a StructDescriptor object does not exist, then you can create one by calling the static StructDescriptor.createDescriptor method. This method requires you to pass in the SQL type name of the Oracle object type and a connection object, as follows:
StructDescriptor structdesc = StructDescriptor.createDescriptor (sql_type_name, connection);

The sql_type_name parameter is a Java String containing the name of the Oracle object type, such as EMPLOYEE, and connection is the connection object. Once you have your StructDescriptor object for the Oracle object type, you can construct the STRUCT object. To do this, provide the Connection object, the StructDescriptor object, and an array of Java objects containing the attributes you want the STRUCT to contain. The following constructors of STRUCT are available:
STRUCT(Connection conn, java.sql.StructDescriptor structDesc, Object[] attributes) STRUCT(Connection conn, java.sql.StructDescriptor structDesc, java.util.Map map)

The structDesc parameter is the StructDescriptor object created previously and conn is your Connection object. The attributes can be passed as an array of java.lang.Object or as a java.util.Map object. The following code illustrates the use of the constructor that takes an Object array:
... Object[] attributes = {"attribute1", null}; STRUCT struct = new STRUCT(connection, structDescriptor, attributes); ...

The following code illustrates the use of the constructor that takes a Map object:

Working with Oracle Object Types 13-5

Creating and Using Custom Object Classes for Oracle Objects

... HashMap map = new HashMap(1); map.put("A1","attribute1"); STRUCT struct = new STRUCT(connection, structDescriptor, map); ...

Binding STRUCT Objects into Statements

To bind an oracle.sql.STRUCT object to a prepared statement or callable statement, you can either use the standard setObject method (specifying the type code), or cast the statement object to an Oracle statement type and use the Oracle extension setOracleObject method. For example:
PreparedStatement ps= conn.prepareStatement("text_of_prepared_statement"); STRUCT mySTRUCT = new STRUCT (...); ps.setObject(1, mySTRUCT, Types.STRUCT);

or:
PreparedStatement ps= conn.prepareStatement("text_of_prepared_statement"); STRUCT mySTRUCT = new STRUCT (...); ((OraclePreparedStatement)ps).setOracleObject(1, mySTRUCT);

STRUCT Automatic Attribute Buffering

Oracle JDBC driver furnishes public methods to enable and disable buffering of STRUCT attributes.
See Also:

"ARRAY Automatic Element Buffering" on page 16-4

The following methods are included with the oracle.sql.STRUCT class:

public void setAutoBuffering(boolean enable) public boolean getAutoBuffering()

The setAutoBuffering(boolean) method enables or disables auto-buffering. The getAutoBuffering method returns the current auto-buffering mode. By default, auto-buffering is disabled. It is advisable to enable auto-buffering in a JDBC application when the STRUCT attributes will be accessed more than once by the getAttributes and getArray methods, presuming the ARRAY data is able to fit into the Java Virtual Machine (JVM) memory without overflow.
Note:

Buffering the converted attributes may cause the JDBC application to consume a significant amount of memory.

When you enable auto-buffering, the oracle.sql.STRUCT object keeps a local copy of all the converted attributes. This data is retained so that subsequent access of this information does not require going through the data format conversion process.

Creating and Using Custom Object Classes for Oracle Objects

If you want to create custom object classes for your Oracle objects, then you must define entries in the type map that specify the custom object classes that the drivers will instantiate for the corresponding Oracle objects.
13-6 Oracle Database JDBC Developer's Guide and Reference

Creating and Using Custom Object Classes for Oracle Objects

You must also provide a way to create and populate instances of the custom object class from the Oracle object and its attribute data. The driver must be able to read from a custom object class and write to it. In addition, the custom object class can provide getXXX and setXXX methods corresponding to the attributes of the Oracle object, although this is not necessary. To create and populate the custom classes and provide these read/write capabilities, you can choose between the following interfaces:

The JDBC standard SQLData interface The ORAData and ORADataFactory interfaces provided by Oracle

The custom object class you create must implement one of these interfaces. The ORAData interface can also be used to implement the custom reference class corresponding to the custom object class. However, if you are using the SQLData interface, then you can use only weak reference types in Java, such as java.sql.Ref or oracle.sql.REF. The SQLData interface is for mapping SQL objects only. As an example, assume you have an Oracle object type, EMPLOYEE, in the database that consists of two attributes: Name, which is of the CHAR type and EmpNum, which is of the NUMBER type. You use the type map to specify that the EMPLOYEE object should map to a custom object class that you call JEmployee. You can implement either the SQLData or ORAData interface in the JEmployee class. You can create custom object classes yourself, but the most convenient way to create them is to use the Oracle JPublisher utility to create them for you. JPublisher supports the standard SQLData interface as well as the Oracle-specific ORAData interface, and is able to generate classes that implement either one.
See Also: "Using JPublisher to Create Custom Object Classes" on page 13-30 and "Object-Type Inheritance" on page 13-19

This section covers the following topics:

Relative Advantages of ORAData versus SQLData Understanding Type Maps for SQLData Implementations Creating Type Map and Defining Mappings for a SQLData Implementation Understanding the SQLData Interface Reading and Writing Data with a SQLData Implementation Understanding the ORAData Interface Reading and Writing Data with a ORAData Implementation Additional Uses for ORAData The Deprecated CustomDatum Interface

Relative Advantages of ORAData versus SQLData

In deciding which of the two interface implementations to use, you need to consider the advantages of ORAData and SQLData. The SQLData interface is for mapping SQL objects only. The ORAData interface is more flexible, enabling you to map SQL objects as well as any other SQL type for which you want to customize processing. You can create a ORAData object from any data type found in Oracle Database. This could be useful, for example, for serializing RAW data in Java.

Working with Oracle Object Types 13-7

Creating and Using Custom Object Classes for Oracle Objects

Advantages of ORAData The advantages of ORAData are:

It does not require an entry in the type map for the Oracle object. It has awareness of Oracle extensions. You can construct an ORAData from an oracle.sql.STRUCT. This is more efficient because it avoids unnecessary conversions to native Java types. You can obtain the corresponding Datum object from the ORAData object, using the toDatum method. It provides better performance. ORAData works directly with Datum types, which is the internal format used by the driver to hold Oracle objects.

Advantages of SQLData SQLData is a JDBC standard that makes your code portable.

Understanding Type Maps for SQLData Implementations

If you use the SQLData interface in a custom object class, then you must create type map entries that specify the custom object class to use in mapping the Oracle object type to Java. You can either use the default type map of the connection object or a type map that you specify when you retrieve the data from the result set. The getObject method of the ResultSet interface has a signature that lets you specify a type map. You can use either of the following:
rs.getObject(int columnIndex); rs.getObject(int columnIndex, Map map);

See Also: "Creating and Using Custom Object Classes for Oracle Objects" on page 13-6

When using a SQLData implementation, if you do not include a type map entry, then the object will map to the oracle.sql.STRUCT class by default. ORAData implementations, by contrast, have their own mapping functionality so that a type map entry is not required. When using an ORAData implementation, use the Oracle getORAData method instead of the standard getObject method. The type map relates a Java class to the SQL type name of an Oracle object. This one-to-one mapping is stored in a hash table as a keyword-value pair. When you read data from an Oracle object, the JDBC driver considers the type map to determine which Java class to use to materialize the data from the Oracle object type. When you write data to an Oracle object, the JDBC driver gets the SQL type name from the Java class by calling the getSQLTypeName method of the SQLData interface. The actual conversion between SQL and Java is performed by the driver. The attributes of the Java class that corresponds to an Oracle object can use either Java native types or Oracle native types to store attributes.

Creating Type Map and Defining Mappings for a SQLData Implementation

When using a SQLData implementation, the JDBC applications programmer is responsible for providing a type map, which must be an instance of a class that implements the standard java.util.Map interface.

13-8 Oracle Database JDBC Developer's Guide and Reference

Creating and Using Custom Object Classes for Oracle Objects

You have the option of creating your own class to accomplish this, but the standard java.util.Hashtable class meets the requirement. Hashtable and other classes used for type maps implement a put method that takes keyword-value pairs as input, where each key is a fully qualified SQL type name and the corresponding value is an instance of a specified Java class. A type map is associated with a connection instance. The standard java.sql.Connection interface and the Oracle-specific oracle.jdbc.OracleConnection interface include a getTypeMap method. Both return a Map object. This section covers the following topics:

Adding Entries to an Existing Type Map Creating a New Type Map Materializing Object Types not Specified in the Type Map

Adding Entries to an Existing Type Map

When a connection instance is first established, the default type map is empty. You must populate it. Perform the following general steps to add entries to an existing type map:
1.

Use the getTypeMap method of your OracleConnection object to return the type map object of the connection. The getTypeMap method returns a java.util.Map object. For example, presuming an OracleConnection instance oraconn:
java.util.Map myMap = oraconn.getTypeMap();

Note: If the type map in the OracleConnection instance has not been initialized, then the first call to getTypeMap returns an empty map.
2.

Use the put method of the type map to add map entries. The put method takes two arguments: a SQL type name string and an instance of a specified Java class that you want to map to.
myMap.put(sqlTypeName, classObject);

The sqlTypeName is a string that represents the fully qualified name of the SQL type in the database. The classObject is the Java class object to which you want to map the SQL type. Get the class object with the Class.forName method, as follows:
myMap.put(sqlTypeName, Class.forName(className));

For example, if you have a PERSON SQL data type defined in the CORPORATE database schema, then map it to a Person Java class defined as Person with this statement:
myMap.put("CORPORATE.PERSON", Class.forName("Person"));

The map has an entry that maps the PERSON SQL data type in the CORPORATE database to the Person Java class.

Working with Oracle Object Types 13-9

Creating and Using Custom Object Classes for Oracle Objects

Note:

SQL type names in the type map must be all uppercase, because that is how Oracle Database stores SQL names.

Creating a New Type Map

Perform the following general steps to create a new type map. This example uses an instance of java.util.Hashtable, which extends java.util.Dictionary and implements java.util.Map.
1.

Create a new type map object.

Hashtable newMap = new Hashtable();

2.

Use the put method of the type map object to add entries to the map. For example, if you have an EMPLOYEE SQL type defined in the CORPORATE database, then you can map it to an Employee class object defined by Employee.java, as follows:
newMap.put("CORPORATE.EMPLOYEE", class.forName("Employee"));

3.

When you finish adding entries to the map, use the setTypeMap method of the OracleConnection object to overwrite the existing type map of the connection. For example:
oraconn.setTypeMap(newMap);

In this example, setTypeMap overwrites the original map of the oraconn connection object with newMap.
Note:

The default type map of a connection instance is used when mapping is required but no map name is specified, such as for a result set getObject call that does not specify the map as input.

Materializing Object Types not Specified in the Type Map

If you do not provide a type map with an appropriate entry when using a getObject call, then the JDBC driver will materialize an Oracle object as an instance of the oracle.sql.STRUCT class. If the Oracle object type contains embedded objects and they are not present in the type map, then the driver will materialize the embedded objects as instances of oracle.sql.STRUCT as well. If the embedded objects are present in the type map, then a call to the getAttributes method will return embedded objects as instances of the specified Java classes from the type map.

Understanding the SQLData Interface

One of the choices in making an Oracle object and its attribute data available to Java applications is to create a custom object class that implements the SQLData interface. Note that if you use this interface, you must supply a type map that specifies the Oracle object types in the database and the names of the corresponding custom object classes that you will create for them. The SQLData interface defines methods that translate between SQL and Java for Oracle database objects. Standard JDBC provides a SQLData interface and companion SQLInput and SQLOutput interfaces in the java.sql package. If you create a custom object class that implements SQLData, then you must provide a readSQL method and a writeSQL method, as specified by the SQLData interface.

13-10 Oracle Database JDBC Developer's Guide and Reference

Creating and Using Custom Object Classes for Oracle Objects

The JDBC driver calls your readSQL method to read a stream of data values from the database and populate an instance of your custom object class. Typically, the driver would use this method as part of an OracleResultSet object getObject call. Similarly, the JDBC driver calls your writeSQL method to write a sequence of data values from an instance of your custom object class to a stream that can be written to the database. Typically, the driver would use this method as part of an OraclePreparedStatement object setObject call. Understanding the SQLInput and SQLOutput Interfaces The JDBC driver includes classes that implement the SQLInput and SQLOutput interfaces. It is not necessary to implement the SQLOutput or SQLInput objects. The JDBC drivers will do this for you. The SQLInput implementation is an input stream class, an instance of which is passed to the readSQL method. SQLInput includes a readXXX method for every possible Java type that attributes of an Oracle object may be converted to, such as readObject, readInt, readLong, readFloat, readBlob, and so on. Each readXXX method converts SQL data to Java data and returns it as the result with the corresponding Java type. For example, readInt returns an int. The SQLOutput implementation is an output stream class, an instance of which is passed in to the writeSQL method. SQLOutput includes a writeXXX method for each of these Java types. Each writeXXX method converts Java data to SQL data, taking as input a parameter of the relevant Java type. For example, writeString would take as input a String attribute from your Java class. Implementing readSQL and writeSQL Methods When you create a custom object class that implements SQLData, you must implement the readSQL and writeSQL methods, as described here. You must implement readSQL as follows:
public void readSQL(SQLInput stream, String sql_type_name) throws SQLException

The readSQL method takes as input a SQLInput stream and a string that indicates the SQL type name of the data, that is, the name of the Oracle object type, such as EMPLOYEE. When your Java application calls getObject, the JDBC driver creates a SQLInput stream object and populates it with data from the database. The driver can also determine the SQL type name of the data when it reads it from the database. When the driver calls readSQL, it passes in these parameters.

For each Java data type that maps to an attribute of the Oracle object, readSQL must call the appropriate readXXX method of the SQLInput stream that is passed in. For example, if you are reading EMPLOYEE objects that have an employee name as a CHAR variable and an employee number as a NUMBER variable, then you must have a readString call and a readInt call in your readSQL method. JDBC calls these methods according to the order in which the attributes appear in the SQL definition of the Oracle object type.

The readSQL method takes the data that the readXXX methods read and convert and assigns them to the appropriate fields or elements of a custom object class instance.

You must implement writeSQL as follows:

Working with Oracle Object Types

13-11

Creating and Using Custom Object Classes for Oracle Objects

public void writeSQL(SQLOutput stream) throws SQLException

The writeSQL method takes as input a SQLOutput stream. When your Java application calls setObject, the JDBC driver creates a SQLOutput stream object. When the driver calls writeSQL, it passes in this stream parameter.

For each Java data type that maps to an attribute of the Oracle object, writeSQL must call the appropriate writeXXX method of the SQLOutput stream that is passed in. For example, if you are writing to EMPLOYEE objects that have an employee name as a CHAR variable and an employee number as a NUMBER variable, then you must have a writeString call and a writeInt call in your writeSQL method. These methods must be called according to the order in which attributes appear in the SQL definition of the Oracle object type.

The writeSQL method then writes the data to the SQLOutput stream by calling the writeXXX methods so that it can be sent to the database once you execute the prepared statement.

Reading and Writing Data with a SQLData Implementation

This section describes how to read data from an Oracle object or write data to an Oracle object if your corresponding Java class implements SQLData. Reading SQLData Objects from a Result Set The following text summarizes the steps to read data from an Oracle object into your Java application when you choose the SQLData implementation for your custom object class. These steps assume you have already defined the Oracle object type, created the corresponding custom object class, updated the type map to define the mapping between the Oracle object and the Java class, and defined a statement object stmt.
1.

Query the database to read the Oracle object into a JDBC result set.
ResultSet rs = stmt.executeQuery("SELECT emp_col FROM personnel");

The PERSONNEL table contains one column, EMP_COL, of SQL type EMP_OBJECT. This SQL type is defined in the type map to map to the Java class Employee.
2.

Use the getObject method of your result set to populate an instance of your custom object class with data from one row of the result set. The getObject method returns the user-defined SQLData object because the type map contains an entry for Employee.
if (rs.next()) Employee emp = (Employee)rs.getObject(1);

Note that if the type map did not have an entry for the object, then getObject would return an oracle.sql.STRUCT object. Cast the output to type STRUCT, because the getObject method signature returns the generic java.lang.Object type.
if (rs.next()) STRUCT empstruct = (STRUCT)rs.getObject(1);

The getObject method calls readSQL, which, in turn, calls readXXX from the SQLData interface.
13-12 Oracle Database JDBC Developer's Guide and Reference

Creating and Using Custom Object Classes for Oracle Objects

Note:

If you want to avoid using the defined type map, then use the getSTRUCT method. This method always returns a STRUCT object, even if there is a mapping entry in the type map.

3.

If you have get methods in your custom object class, then use them to read data from your object attributes. For example, if EMPLOYEE has the attributes EmpName of type CHAR and EmpNum of type NUMBER, then provide a getEmpName method that returns a Java String and a getEmpNum method that returns an int value. Then call them in your Java application, as follows:
String empname = emp.getEmpName(); int empnumber = emp.getEmpNum();

Retrieving SQLData Objects from a Callable Statement OUT Parameter Consider you have an OracleCallableStatement instance, ocs, that calls a PL/SQL function GETEMPLOYEE. The program passes an employee number to the function. The function returns the corresponding Employee object. To retrieve this object you do the following:
1.

Prepare an OracleCallableStatement to call the GETEMPLOYEE function, as follows:

OracleCallableStatement ocs = (OracleCallableStatement)conn.prepareCall("{ ? = call GETEMPLOYEE(?) }");

2.

Declare the empnumber as the input parameter to GETEMPLOYEE. Register the SQLData object as the OUT parameter, with the type code OracleTypes.STRUCT. Then, run the statement. This can be done as follows:
ocs.setInt(2, empnumber); ocs.registerOutParameter(1, OracleTypes.STRUCT, "EMP_OBJECT"); ocs.execute();

3.

Use the getObject method to retrieve the employee object. The following code assumes that there is a type map entry to map the Oracle object to the Java type Employee:
Employee emp = (Employee)ocs.getObject(1);

If there is no type map entry, then getObject would return an oracle.sql.STRUCT object. Cast the output to the STRUCT type, because the getObject method returns an instance of the generic java.lang.Object class. This is done as follows:
STRUCT emp = (STRUCT)ocs.getObject(1);

Passing SQLData Objects to a Callable Statement as an IN Parameter Suppose you have a PL/SQL function addEmployee(?) that takes an Employee object as an IN parameter and adds it to the PERSONNEL table. In this example, emp is a valid Employee object.
1.

Prepare an OracleCallableStatement to call the addEmployee(?) function.

OracleCallableStatement ocs = (OracleCallableStatement) conn.prepareCall("{ call addEmployee(?) }");

Working with Oracle Object Types

13-13

Creating and Using Custom Object Classes for Oracle Objects

2.

Use setObject to pass the emp object as an IN parameter to the callable statement. Then, call the statement.
ocs.setObject(1, emp); ocs.execute();

Writing Data to an Oracle Object Using a SQLData Implementation The following text describes the steps in writing data to an Oracle object from your Java application when you choose the SQLData implementation for your custom object class. This description assumes you have already defined the Oracle object type, created the corresponding Java class, and updated the type map to define the mapping between the Oracle object and the Java class.
1.

If you have set methods in your custom object class, then use them to write data from Java variables in your application to attributes of your Java data type object.
emp.setEmpName(empname); emp.setEmpNum(empnumber);

This statement uses the emp object and the empname and empnumber variables assigned in the preceding example.
2.

Prepare a statement that updates an Oracle object in a row of a database table, as appropriate, using the data provided in your Java data type object.
PreparedStatement pstmt = conn.prepareStatement ("INSERT INTO PERSONNEL VALUES (?)");

This assumes conn is your connection object.

3.

Use the setObject method of the prepared statement to bind your Java data type object to the prepared statement.
pstmt.setObject(1, emp);

4.

Run the statement, which updates the database.

pstmt.executeUpdate();

Understanding the ORAData Interface

One of the choices in making an Oracle object and its attribute data available to Java applications is to create a custom object class that implements the oracle.sql.ORAData and oracle.sql.ORADataFactory interfaces. The ORAData and ORADataFactory interfaces are supplied by Oracle and are not a part of the JDBC standard.
Note:

The JPublisher utility supports the generation of classes that implement the ORAData and ORADataFactory interfaces.

Understanding ORAData Features The ORAData interface has the following advantages:

It recognizes Oracle extensions to the JDBC. ORAData uses oracle.sql.Datum types directly.

13-14 Oracle Database JDBC Developer's Guide and Reference

Creating and Using Custom Object Classes for Oracle Objects

It does not require a type map to specify the names of the Java custom classes you want to create. It provides better performance. ORAData works directly with Datum types, the internal format the driver uses to hold Oracle objects.

The ORAData and ORADataFactory interfaces do the following:

The toDatum method of the ORAData class transforms the data into an oracle.sql.* representation. ORADataFactory specifies a create method equivalent to a constructor for your custom object class. It creates and returns an ORAData instance. The JDBC driver uses the create method to return an instance of the custom object class to your Java application or applet. It takes as input an oracle.sql.Datum object and an integer indicating the corresponding SQL type code as specified in the OracleTypes class.

ORAData and ORADataFactory have the following definitions:

public interface ORAData { Datum toDatum (OracleConnection conn) throws SQLException; } public interface ORADataFactory { ORAData create (Datum d, int sql_Type_Code) throws SQLException; }

Where conn represents the Connection object, d represents an object of type oracle.sql.Datum and sql_Type_Code represents the SQL type code of the Datum object. Retrieving and Inserting Object Data The JDBC drivers provide the following methods to retrieve and insert object data as instances of ORAData. You can retrieve the object data in one of the following ways:

Use the following getORAData method of the Oracle-specific OracleResultSet class:

ors.getORAData (int col_index, ORADataFactory factory);

This method takes as input the column index of the data in your result set and a ORADataFactory instance. For example, you can implement a getORAFactory method in your custom object class to produce the ORADataFactory instance to input to getORAData. The type map is not required when using Java classes that implement ORAData.

Use the standard getObject(index, map) method specified by the ResultSet interface to retrieve data as instances of ORAData. In this case, you must have an entry in the type map that identifies the factory class to be used for the given object type and its corresponding SQL type name.

You can insert object data in one of the following ways:

Use the following setORAData method of the Oracle-specific OraclePreparedStatement class:

ops.setORAData (int bind_index, ORAData custom_obj);

Working with Oracle Object Types

13-15

Creating and Using Custom Object Classes for Oracle Objects

This method takes as input the parameter index of the bind variable and the name of the object containing the variable.

Use the standard setObject method specified by the PreparedStatement interface. You can also use this method, in its different forms, to insert ORAData instances without requiring a type map.

The following sections describe the getORAData and setORAData methods. To continue the example of an Oracle object EMPLOYEE, you might have something like the following in your Java application:
ORAData datum = ors.getORAData(1, Employee.getORAFactory());

In this example, ors is an Oracle result set, getORAData is a method in the OracleResultSet class used to retrieve a ORAData object, and the EMPLOYEE is in column 1 of the result set. The static Employee.getORAFactory method will return a ORADataFactory to the JDBC driver. The JDBC driver will call create() from this object, returning to your Java application an instance of the Employee class populated with data from the result set.
Note:

ORAData and ORADataFactory are defined as separate interfaces so that different Java classes can implement them if you wish. To use the ORAData interface, your custom object classes must import oracle.sql.*.

Reading and Writing Data with a ORAData Implementation

This section describes how to read data from an Oracle object or write data to an Oracle object if your corresponding Java class implements ORAData. Reading Data from an Oracle Object Using a ORAData Implementation The following text summarizes the steps in reading data from an Oracle object into your Java application. These steps apply whether you implement ORAData manually or use JPublisher to produce your custom object classes. These steps assume you have already defined the Oracle object type, created the corresponding custom object class or had JPublisher create it for you, and defined a statement object stmt.
1.

Query the database to read the Oracle object into a result set, casting it to an Oracle result set.
OracleResultSet ors = (OracleResultSet)stmt.executeQuery ("SELECT Emp_col FROM PERSONNEL");

Where PERSONNEL is a one-column table. The column name is Emp_col of type Employee_object.
2.

Use the getORAData method of your Oracle result set to populate an instance of your custom object class with data from one row of the result set. The getORAData method returns an oracle.sql.ORAData object, which you can cast to your specific custom object class.
if (ors.next()) Employee emp = (Employee)ors.getORAData(1, Employee.getORAFactory());

13-16 Oracle Database JDBC Developer's Guide and Reference

Creating and Using Custom Object Classes for Oracle Objects

or:
if (ors.next()) ORAData datum = ors.getORAData(1, Employee.getORAFactory());

This example assumes that Employee is the name of your custom object class and ors is the name of your OracleResultSet object. In case you do not want to use getORAData, the JDBC drivers let you use the getObject method of a standard JDBC ResultSet to retrieve ORAData data. However, you must have an entry in the type map that identifies the factory class to be used for the given object type and its corresponding SQL type name. For example, if the SQL type name for your object is EMPLOYEE, then the corresponding Java class is Employee, which will implement ORAData. The corresponding Factory class is EmployeeFactory, which will implement ORADataFactory. Use this statement to declare the EmployeeFactory entry for your type map:
map.put ("EMPLOYEE", Class.forName ("EmployeeFactory"));

Then use the form of getObject where you specify the map object:
Employee emp = (Employee) rs.getObject (1, map);

If the default type map of the connection already has an entry that identifies the factory class to be used for the given object type and its corresponding SQL type name, then you can use this form of getObject:
Employee emp = (Employee) rs.getObject (1); 3.

If you have get methods in your custom object class, then use them to read data from your object attributes into Java variables in your application. For example, if EMPLOYEE has EmpName of type CHAR and EmpNum of type NUMBER, provide a getEmpName method that returns a Java String and a getEmpNum method that returns an integer. Then call them in your Java application as follows:
String empname = emp.getEmpName(); int empnumber = emp.getEmpNum();

Note: Alternatively, you can fetch data using a callable statement object. The OracleCallableStatement class also has a getORAData method.

Writing Data to an Oracle Object Using a ORAData Implementation The following text summarizes the steps in writing data to an Oracle object from your Java application. These steps apply whether you implement ORAData manually or use JPublisher to produce your custom object classes. These steps assume you have already defined the Oracle object type and created the corresponding custom object class.
Note:

The type map is not used when you are performing database INSERT and UPDATE operations.

Working with Oracle Object Types

13-17

Creating and Using Custom Object Classes for Oracle Objects

1.

If you have set methods in your custom object class, then use them to write data from Java variables in your application to attributes of your Java data type object.
emp.setEmpName(empname); emp.setEmpNum(empnumber);

2.

Write an Oracle prepared statement that updates an Oracle object in a row of a database table, as appropriate, using the data provided in your Java data type object.
OraclePreparedStatement opstmt = conn.prepareStatement ("UPDATE PERSONNEL SET Employee = ? WHERE Employee.EmpNum = 28959);

This assumes conn is your Connection object.

3.

Use the setORAData method of the Oracle prepared statement to bind your Java data type object to the prepared statement.
opstmt.setORAData(1, emp);

The setORAData method calls the toDatum method of the custom object class instance to retrieve an oracle.sql.STRUCT object that can be written to the database. In this step you could also use the setObject method to bind the Java data type. For example:
opstmt.setObject(1,emp);

Note: You can use your Java data type objects as either IN or OUT bind variables.

Additional Uses for ORAData

The ORAData interface offers far more flexibility than the SQLData interface. The SQLData interface is designed to let you customize the mapping of only Oracle object types to Java types of your choice. Implementing the SQLData interface lets the JDBC driver populate fields of a custom Java class instance from the original SQL object data, and the reverse, after performing the appropriate conversions between Java and SQL types. The ORAData interface goes beyond supporting the customization of Oracle object types to Java types. It lets you provide a mapping between Java object types and any SQL type supported by the oracle.sql package. It may be useful to provide custom Java classes to wrap oracle.sql.* types and perhaps implement customized conversions or functionality as well. The following are some possible scenarios:

Performing encryption and decryption or validation of data Performing logging of values that have been read or are being written Parsing character columns, such as character fields containing URL information, into smaller components Mapping character strings into numeric constants Making data into more desirable Java formats, such as mapping a DATE field to java.util.Date format

13-18 Oracle Database JDBC Developer's Guide and Reference

Object-Type Inheritance

Customizing data representation, for example, data in a table column is in feet but you want it represented in meters after it is selected Serializing and deserializing Java objects

For example, use ORAData to store instances of Java objects that do not correspond to a particular SQL object type in the database in columns of SQL type RAW. The create method in ORADataFactory would have to implement a conversion from an object of type oracle.sql.RAW to the desired Java object. The toDatum method in ORAData would have to implement a conversion from the Java object to an oracle.sql.RAW object. This can be done, for example, by using Java serialization. Upon retrieval, the JDBC driver transparently retrieves the raw bytes of data in the form of an oracle.sql.RAW and calls the create method of ORADataFactory to convert the oracle.sql.RAW object to the desired Java class. When you insert the Java object into the database, you can simply bind it to a column of type RAW to store it. The driver transparently calls the ORAData.toDatum method to convert the Java object to an oracle.sql.RAW object. This object is then stored in a column of type RAW in the database. Support for the ORAData interfaces is also highly efficient because the conversions are designed to work using oracle.sql.* formats, which happen to be the internal formats used by the JDBC drivers. Moreover, the type map, which is necessary for the SQLData interface, is not required when using Java classes that implement ORAData.
See Also:

"Understanding the ORAData Interface" on page 13-14

The Deprecated CustomDatum Interface

After the oracle.jdbc interfaces were introduced in Oracle9i Database as an alternative to the oracle.jdbc.driver classes, the oracle.sql.CustomDatum and oracle.sql.CustomDatumFactory interfaces, formerly used to access customized objects, were deprecated. Oracle recommends you use the new interfaces, oracle.sql.ORAData and oracle.sql.ORADataFactory.

Object-Type Inheritance
Object-type inheritance allows a new object type to be created by extending another object type. The new object type is then a subtype of the object type from which it extends. The subtype automatically inherits all the attributes and methods defined in the supertype. The subtype can add attributes and methods and overload or override methods inherited from the supertype. Object-type inheritance introduces substitutability. Substitutability is the ability of a slot declared to hold a value of type T in addition to any subtype of type T. Oracle JDBC drivers handle substitutability transparently. A database object is returned with its most specific type without losing information. For example, if the STUDENT_T object is stored in a PERSON_T slot, Oracle JDBC driver returns a Java object that represents the STUDENT_T object. This section covers the following topics:

Creating Subtypes Implementing Customized Classes for Subtypes Retrieving Subtype Objects Creating Subtype Objects

Working with Oracle Object Types

13-19

Object-Type Inheritance

Sending Subtype Objects Accessing Subtype Data Fields Inheritance Metadata Methods

Creating Subtypes
Create custom object classes if you want to have Java classes that explicitly correspond to the Oracle object types. If you have a hierarchy of object types, you may want a corresponding hierarchy of Java classes. The most common way to create a database subtype in JDBC is to run a SQL CREATE TYPE command using the execute method of the java.sql.Statement interface. For example, you want to create a type inheritance hierarchy for:
PERSON_T | STUDENT_T | PARTTIMESTUDENT_T

The JDBC code for this can be as follows:

Statement s = conn.createStatement(); s.execute ("CREATE TYPE Person_T (SSN NUMBER, name VARCHAR2(30), address VARCHAR2(255))"); s.execute ("CREATE TYPE Student_T UNDER Person_t (deptid NUMBER, major VARCHAR2(100))"); s.execute ("CREATE TYPE PartTimeStudent_t UNDER Student_t (numHours NUMBER)");

In the following code, the foo member procedure in type ST is overloaded and the member procedure print overwrites the copy it inherits from type T.
CREATE TYPE T AS OBJECT (..., MEMBER PROCEDURE foo(x NUMBER), MEMBER PROCEDURE Print(), ... NOT FINAL; CREATE TYPE ST UNDER T (..., MEMBER PROCEDURE foo(x DATE), OVERRIDING MEMBER PROCEDURE Print(), STATIC FUNCTION bar(...) ... ... );

<-- overload "foo" <-- override "print"

Once the subtypes have been created, they can be used as both columns of a base table as well as attributes of a object type.
See Also:

Oracle Database Object-Relational Developer's Guide

Implementing Customized Classes for Subtypes

In most cases, a customized Java class represents a database object type. When you create a customized Java class for a subtype, the Java class can either mirror the database object type hierarchy or not. You can use either the ORAData or SQLData solution in creating classes to map to the hierarchy of object types. This section covers the following topics:
13-20 Oracle Database JDBC Developer's Guide and Reference

Object-Type Inheritance

Use of ORAData for Type Inheritance Hierarchy Use of SQLData for Type Inheritance Hierarchy JPublisher Utility

Use of ORAData for Type Inheritance Hierarchy

Customized mapping where Java classes implement the oracle.sql.ORAData interface is the recommended mapping. ORAData mapping requires the JDBC application to implement the ORAData and ORADataFactory interfaces. The class implementing the ORADataFactory interface contains a factory method that produces objects. Each object represents a database object. The hierarchy of the class implementing the ORAData interface can mirror the database object type hierarchy. For example, the Java classes mapping to PERSON_T and STUDENT_T are as follows: Person.java using ORAData Code for the Person.java class which implements the ORAData and ORADataFactory interfaces:
class Person implements ORAData, ORADataFactory { static final Person _personFactory = new Person(); public NUMBER ssn; public CHAR name; public CHAR address; public static ORADataFactory getORADataFactory() { return _personFactory; } public Person () {} public Person(NUMBER ssn, CHAR name, CHAR address) { this.ssn = ssn; this.name = name; this.address = address; } public Datum toDatum(OracleConnection c) throws SQLException { StructDescriptor sd = StructDescriptor.createDescriptor("SCOTT.PERSON_T", c); Object [] attributes = { ssn, name, address }; return new STRUCT(sd, c, attributes); } public ORAData create(Datum d, int sqlType) throws SQLException { if (d == null) return null; Object [] attributes = ((STRUCT) d).getOracleAttributes(); return new Person((NUMBER) attributes[0], (CHAR) attributes[1], (CHAR) attributes[2]); } } Working with Oracle Object Types 13-21

Object-Type Inheritance

Student.java extending Person.java Code for the Student.java class, which extends the Person.java class:
class Student extends Person { static final Student _studentFactory = new Student (); public NUMBER deptid; public CHAR major; public static ORADataFactory getORADataFactory() { return _studentFactory; } public Student () {} public Student (NUMBER ssn, CHAR name, CHAR address, NUMBER deptid, CHAR major) { super (ssn, name, address); this.deptid = deptid; this.major = major; } public Datum toDatum(OracleConnection c) throws SQLException { StructDescriptor sd = StructDescriptor.createDescriptor("SCOTT.STUDENT_T", c); Object [] attributes = { ssn, name, address, deptid, major }; return new STRUCT(sd, c, attributes); } public CustomDatum create(Datum d, int sqlType) throws SQLException { if (d == null) return null; Object [] attributes = ((STRUCT) d).getOracleAttributes(); return new Student((NUMBER) attributes[0], (CHAR) attributes[1], (CHAR) attributes[2], (NUMBER) attributes[3], (CHAR) attributes[4]); } }

Customized classes that implement the ORAData interface do not have to mirror the database object type hierarchy. For example, you could have declared the Student class without a superclass. In this case, Student would contain fields to hold the inherited attributes from PERSON_T as well as the attributes declared by STUDENT_T. ORADataFactory Implementation The JDBC application uses the factory class in querying the database to return instances of Person or its subclasses, as in the following example:
ResultSet rset = stmt.executeQuery ("select person from tab1"); while (rset.next()) {

13-22 Oracle Database JDBC Developer's Guide and Reference

Object-Type Inheritance

Object s = rset.getORAData (1, PersonFactory.getORADataFactory()); ... }

A class implementing the ORADataFactory interface should be able to produce instances of the associated custom object type, as well as instances of any subtype, or at least all the types you expect to support. In the following example, the PersonFactory.getORADataFactory method returns a factory that can handle PERSON_T, STUDENT_T, and PARTTIMESTUDENT_T objects, by returning person, student, or parttimestudent Java instances.
class PersonFactory implements ORADataFactory { static final PersonFactory _factory = new PersonFactory (); public static ORADataFactory getORADataFactory() { return _factory; } public ORAData create(Datum d, int sqlType) throws SQLException { STRUCT s = (STRUCT) d; if (s.getSQLTypeName ().equals ("SCOTT.PERSON_T")) return Person.getORADataFactory ().create (d, sqlType); else if (s.getSQLTypeName ().equals ("SCOTT.STUDENT_T")) return Student.getORADataFactory ().create(d, sqlType); else if (s.getSQLTypeName ().equals ("SCOTT.PARTTIMESTUDENT_T")) return ParttimeStudent.getORADataFactory ().create(d, sqlType); else return null; } }

The following example assumes a table tabl1, such as the following:

CREATE TABLE tabl1 (idx NUMBER, person PERSON_T); INSERT INTO tabl1 VALUES (1, PERSON_T (1000, 'Scott', '100 Oracle Parkway')); INSERT INTO tabl1 VALUES (2, STUDENT_T (1001, 'Peter', '200 Oracle Parkway', 101, 'CS')); INSERT INTO tabl1 VALUES (3, PARTTIMESTUDENT_T (1002, 'David', '300 Oracle Parkway', 102, 'EE'));

Use of SQLData for Type Inheritance Hierarchy

The customized classes that implement the java.sql.SQLData interface can mirror the database object type hierarchy. The readSQL and writeSQL methods of a subclass typically call the corresponding superclass methods to read or write the superclass attributes before reading or writing the subclass attributes. For example, the Java classes mapping to PERSON_T and STUDENT_T are as follows: Person.java using SQLData Code for the Person.java class, which implements the SQLData interface:
import java.sql.*; public class Person implements SQLData { private String sql_type; Working with Oracle Object Types 13-23

Object-Type Inheritance

public int ssn; public String name; public String address; public Person () {} public String getSQLTypeName() throws SQLException { return sql_type; } public void readSQL(SQLInput stream, String typeName) throws SQLException { sql_type = typeName; ssn = stream.readInt(); name = stream.readString(); address = stream.readString(); } public void writeSQL(SQLOutput stream) throws SQLException { stream.writeInt (ssn); stream.writeString (name); stream.writeString (address); } }

Student.java extending Student.java Code for the Student.java class, which extends the Person.java class:
import java.sql.*; public class Student extends Person { private String sql_type; public int deptid; public String major; public Student () { super(); } public String getSQLTypeName() throws SQLException { return sql_type; } public void readSQL(SQLInput stream, String typeName) throws SQLException { super.readSQL (stream, typeName); // read supertype attributes sql_type = typeName; deptid = stream.readInt(); major = stream.readString(); } public void writeSQL(SQLOutput stream) throws SQLException { super.writeSQL (stream); // write supertype // attributes stream.writeInt (deptid); stream.writeString (major); } }

Although not required, it is recommended that the customized classes, which implement the SQLData interface, mirror the database object type hierarchy. For example, you could have declared the Student class without a superclass. In this
13-24 Oracle Database JDBC Developer's Guide and Reference

Object-Type Inheritance

case, Student would contain fields to hold the inherited attributes from PERSON_T as well as the attributes declared by STUDENT_T. Student.java using SQLData Code for the Student.java class, which does not extend the Person.java class, but implements the SQLData interface directly:
import java.sql.*; public class Student implements SQLData { private String sql_type; public public public public public int ssn; String name; String address; int deptid; String major;

public Student () {} public String getSQLTypeName() throws SQLException { return sql_type; } public void readSQL(SQLInput stream, String typeName) throws SQLException { sql_type = typeName; ssn = stream.readInt(); name = stream.readString(); address = stream.readString(); deptid = stream.readInt(); major = stream.readString(); } public void writeSQL(SQLOutput stream) throws SQLException { stream.writeInt (ssn); stream.writeString (name); stream.writeString (address); stream.writeInt (deptid); stream.writeString (major); } }

JPublisher Utility
Even though you can manually create customized classes that implement the SQLData, ORAData, and ORADataFactory interfaces, it is recommended that you use Oracle JPublisher to automatically generate these classes. The customized classes generated by Oracle JPublisher that implement the SQLData, ORAData, and ORADataFactory interfaces, can mirror the inheritance hierarchy.
See Also:

"Using JPublisher to Create Custom Object Classes" on page 13-30 Oracle Database JPublisher User's Guide

Working with Oracle Object Types

13-25

Object-Type Inheritance

Retrieving Subtype Objects

In a typical JDBC application, a subtype object is returned as one of the following:

A query result A PL/SQL OUT parameter A type attribute

You can use either the default mapping or the SQLData mapping or the ORAData mapping to retrieve a subtype. Using Default Mapping By default, a database object is returned as an instance of the oracle.sql.STRUCT class. This instance may represent an object of either the declared type or subtype of the declared type. If the STRUCT class represents a subtype object in the database, then it contains the attributes of its supertype as well as those defined in the subtype. Oracle JDBC driver returns database objects in their most specific type. The JDBC application can use the getSQLTypeName method of the STRUCT class to determine the SQL type of the STRUCT object. The following code shows this:
// tab1.person column can store PERSON_T, STUDENT_T and PARTIMESTUDENT_T objects ResultSet rset = stmt.executeQuery ("select person from tab1"); while (rset.next()) { oracle.sql.STRUCT s = (oracle.sql.STRUCT) rset.getObject(1); if (s != null) System.out.println (s.getSQLTypeName()); // print out the type name which // may be SCOTT.PERSON_T, SCOTT.STUDENT_T or SCOTT.PARTTIMESTUDENT_T }

Using SQLData Mapping With SQLData mapping, the JDBC driver returns the database object as an instance of the class implementing the SQLData interface. To use SQLData mapping in retrieving database objects, do the following:
1. 2. 3.

Implement the container classes that implement the SQLData interface for the desired object types. Populate the connection type map with entries that specify what custom Java type corresponds to each Oracle object type. Use the getObject method to access the SQL object values. The JDBC driver checks the type map for an entry match. If one exists, then the driver returns the database object as an instance of the class implementing the SQLData interface.

The following code shows the whole SQLData customized mapping process:
// The JDBC application developer implements Person.java for PERSON_T, // Student.java for STUDENT_T // and ParttimeStudent.java for PARTTIMESTUDEN_T. Connection conn = ...; // make a JDBC connection

// obtains the connection typemap java.util.Map map = conn.getTypeMap ();

13-26 Oracle Database JDBC Developer's Guide and Reference

Object-Type Inheritance

// populate the type map map.put ("SCOTT.PERSON_T", Class.forName ("Person")); map.put ("SCOTT.STUDENT_T", Class.forName ("Student")); map.put ("SCOTT.PARTTIMESTUDENT_T", Class.forName ("ParttimeStudent")); // tab1.person column can store PERSON_T, STUDENT_T and PARTTIMESTUDENT_T objects ResultSet rset = stmt.executeQuery ("select person from tab1"); while (rset.next()) { // "s" is instance of Person, Student or ParttimeStudent Object s = rset.getObject(1); if (s != null) { if (s instanceof Person) System.out.println ("This is a Person"); else if (s instanceof Student) System.out.println ("This is a Student"); else if (s instanceof ParttimeStudent) System.out.pritnln ("This is a PartimeStudent"); else System.out.println ("Unknown type"); } }

The JDBC drivers check the connection type map for each call to the following:

getObject method of the java.sql.ResultSet and java.sql.CallableStatement interfaces getAttribute method of the java.sql.Struct interface getArray method of the java.sql.Array interface getValue method of the oracle.sql.REF interface

Using ORAData Mapping With ORAData mapping, the JDBC driver returns the database object as an instance of the class implementing the ORAData interface. Oracle JDBC driver needs to be informed of what Java class is mapped to the Oracle object type. The following are the two ways to inform Oracle JDBC drivers:

The JDBC application uses the getORAData(int idx, ORADataFactory f) method to access database objects. The second parameter of the getORAData method specifies an instance of the factory class that produces the customized class. The getORAData method is available in the OracleResultSet and OracleCallableStatement classes. The JDBC application populates the connection type map with entries that specify what custom Java type corresponds to each Oracle object type. The getObject method is used to access the Oracle object values.

The second approach involves the use of the standard getObject method. The following code example demonstrates the first approach:
// tab1.person column can store both PERSON_T and STUDENT_T objects ResultSet rset = stmt.executeQuery ("select person from tab1"); while (rset.next()) { Object s = rset.getORAData (1, PersonFactory.getORADataFactory()); if (s != null)

Working with Oracle Object Types

13-27

Object-Type Inheritance

{ if (s instanceof Person) System.out.println ("This is a Person"); else if (s instanceof Student) System.out.println ("This is a Student"); else if (s instanceof ParttimeStudent) System.out.pritnln ("This is a PartimeStudent"); else System.out.println ("Unknown type"); } }

Creating Subtype Objects

There are cases where JDBC applications create database subtype objects with JDBC drivers. These objects are sent either to the database as bind variables or are used to exchange information within the JDBC application. With customized mapping, the JDBC application creates either SQLData- or ORAData-based objects, depending on the approach you choose, to represent database subtype objects. With default mapping, the JDBC application creates STRUCT objects to represent database subtype objects. All the data fields inherited from the supertype as well as all the fields defined in the subtype must have values. The following code demonstrates this:
Connection conn = ... // make a JDBC connection StructDescriptor desc = StructDescriptor.createDescriptor ("SCOTT.PARTTIMESTUDENT", conn); Object[] attrs = { new Integer(1234), "Scott", "500 Oracle Parkway", // data fields defined in // PERSON_T new Integer(102), "CS", // data fields defined in // STUDENT_T new Integer(4) // data fields defined in // PARTTIMESTUDENT_T }; STRUCT s = new STRUCT (desc, conn, attrs);

s is initialized with data fields inherited from PERSON_T and STUDENT_T, and data fields defined in PARTTIMESTUDENT_T.

Sending Subtype Objects

In a typical JDBC application, a Java object that represents a database object is sent to the databases as one of the following:

A data manipulation language (DML) bind variable A PL/SQL IN parameter An object type attribute value

The Java object can be an instance of the STRUCT class or an instance of the class implementing either the SQLData or ORAData interface. Oracle JDBC driver will convert the Java object into the linearized format acceptable to the database SQL engine. Binding a subtype object is the same as binding a standard object.

13-28 Oracle Database JDBC Developer's Guide and Reference

Object-Type Inheritance

Accessing Subtype Data Fields

While the logic to access subtype data fields is part of the customized class, this logic for default mapping is defined in the JDBC application itself. The database objects are returned as instances of the oracle.sql.STRUCT class. The JDBC application needs to call one of the following access methods in the STRUCT class to access the data fields:

Object[] getAttribute() oracle.sql.Datum[] getOracleAttribute()

Subtype Data Fields from the getAttribute Method The getAttribute method of the java.sql.Struct interface is used in JDBC 2.0 to access object data fields. This method returns a java.lang.Object array, where each array element represents an object attribute. You can determine the individual element type by referencing the corresponding attribute type in the JDBC conversion matrix, as listed in Table 41. For example, a SQL NUMBER attribute is converted to a java.math.BigDecimal object. The getAttribute method returns all the data fields defined in the supertype of the object type as well as data fields defined in the subtype. The supertype data fields are listed first followed by the subtype data fields. Subtype Data Fields from the getOracleAttribute Method The getOracleAttribute method is an Oracle extension method and is more efficient than the getAttribute method. The getOracleAttribute method returns an oracle.sql.Datum array to hold the data fields. Each element in the oracle.sql.Datum array represents an attribute. You can determine the individual element type by referencing the corresponding attribute type in the Oracle conversion matrix, as listed in Table 41. For example, a SQL NUMBER attribute is converted to an oracle.sql.NUMBER object. The getOracleAttribute method returns all the attributes defined in the supertype of the object type, as well as attributes defined in the subtype. The supertype data fields are listed first followed by the subtype data fields. The following code shows the use of the getAttribute method:
// tab1.person column can store PERSON_T, STUDENT_T and PARTIMESTUDENT_T objects ResultSet rset = stmt.executeQuery ("select person from tab1"); while (rset.next()) { oracle.sql.STRUCT s = (oracle.sql.STRUCT) rset.getObject(1); if (s != null) { String sqlname = s.getSQLTypeName(); Object[] attrs = s.getAttribute(); if (sqlname.equals ("SCOTT.PERSON") { System.out.println ("ssn="+((BigDecimal)attrs[0]).intValue()); System.out.println ("name="+((String)attrs[1])); System.out.println ("address="+((String)attrs[2])); } else if (sqlname.equals ("SCOTT.STUDENT")) { System.out.println ("ssn="+((BigDecimal)attrs[0]).intValue()); System.out.println ("name="+((String)attrs[1])); System.out.println ("address="+((String)attrs[2])); System.out.println ("deptid="+((BigDecimal)attrs[3]).intValue());

Working with Oracle Object Types

13-29

Using JPublisher to Create Custom Object Classes

System.out.println ("major="+((String)attrs[4])); } else if (sqlname.equals ("SCOTT.PARTTIMESTUDENT")) { System.out.println ("ssn="+((BigDecimal)attrs[0]).intValue()); System.out.println ("name="+((String)attrs[1])); System.out.println ("address="+((String)attrs[2])); System.out.println ("deptid="+((BigDecimal)attrs[3]).intValue()); System.out.println ("major="+((String)attrs[4])); System.out.println ("numHours="+((BigDecimal)attrs[5]).intValue()); } else throw new Exception ("Invalid type name: "+sqlname); } } rset.close (); stmt.close (); conn.close ();

Inheritance Metadata Methods

Oracle JDBC drivers provide a set of metadata methods to access inheritance properties. The inheritance metadata methods are defined in the oracle.sql.StructDescriptor and oracle.jdbc.StructMetaData classes. The StructMetaData class provides inheritance metadata methods for subtype attributes. The getMetaData method of the StructDescriptor class returns an instance of StructMetaData of the type. The StructMetaData class contains the following inheritance metadata methods:

Using JPublisher to Create Custom Object Classes

A convenient way to create custom object classes, as well as other kinds of custom Java classes, is to use the Oracle JPublisher utility. It generates a full definition for a custom Java class, which you can instantiate to hold the data from an Oracle object. JPublisher-generated classes include methods to convert data from SQL to Java and from Java to SQL, as well as getter and setter methods for the object attributes. This section covers the following topics:

JPublisher Functionality JPublisher Type Mappings

See Also:

Oracle Database JPublisher User's Guide.

JPublisher Functionality
You can direct JPublisher to create custom object classes that implement either the SQLData interface or the ORAData interface, according to how you set the JPublisher type mappings. If you use the ORAData interface, then JPublisher will also create a custom reference class to map to object references for the Oracle object type. If you use the SQLData interface, then JPublisher will not produce a custom reference class. You would use standard java.sql.Ref instances instead. If you want additional functionality, you can subclass the custom object class and add features as desired. When you run JPublisher, there is a command-line option for
13-30 Oracle Database JDBC Developer's Guide and Reference

Using JPublisher to Create Custom Object Classes

specifying both a generated class name and the name of the subclass you will implement. For the SQL-Java mapping to work properly, JPublisher must know the subclass name, which is incorporated into some of the functionality of the generated class.
Note: Hand-editing the JPublisher-generated class, instead of subclassing it, is not recommended. If you hand-edit this class and later have to re-run JPublisher for some reason, you would have to re-implement your changes.

JPublisher Type Mappings

JPublisher offers various choices for how to map user-defined types and their attribute types between SQL and Java. This section lists categories of SQL types and the mapping options available for each category. Categories of SQL Types JPublisher categorizes SQL types into the following groups, with corresponding JPublisher options as specifies:

User-defined types (UDT) This includes Oracle objects, references, and collections. You use the JPublisher -usertypes option to specify the type-mapping implementation for UDTs, either a standard SQLData implementation or an Oracle-specific ORAData implementation.

Numeric types This includes anything stored in the database as the NUMBER SQL type. You use the JPublisher -numbertypes option to specify type-mapping for numeric types.

Large object (LOB) types This includes the SQL types, BLOB and CLOB. You use the JPublisher -lobtypes option to specify type-mapping for LOB types.

Built-in types This includes anything stored in the database as a SQL type not covered by the preceding categories. For example, CHAR, VARCHAR2, LONG, and RAW. You use the JPublisher -builtintypes option to specify type-mapping for built-in types.

Type-Mapping Modes JPublisher defines the following type-mapping modes, two of which apply to numeric types only:

JDBC mapping (setting jdbc) Uses standard default mappings between SQL types and Java native types. For a custom object class, uses a SQLData implementation.

Oracle mapping (setting oracle) Uses corresponding oracle.sql types to map to SQL types. For a custom object, reference, or collection class, uses a ORAData implementation.

Object-JDBC mapping (setting objectjdbc) Is an extension of the JDBC mapping. Where relevant, object-JDBC mapping uses numeric object types from the standard java.lang package, such as
Working with Oracle Object Types 13-31

Using JPublisher to Create Custom Object Classes

java.lang.Integer, Float, and Double, instead of primitive Java types, such as int, float, and double. The java.lang types are nullable, while the primitive types are not.

BigDecimal mapping (setting bigdecimal) Uses java.math.BigDecimal to map to all numeric attributes. This is appropriate if you are dealing with large numbers but do not want to map to the oracle.sql.NUMBER class.
Note:

Using BigDecimal mapping can significantly degrade performance.

Mapping the Oracle object type to Java Use the JPublisher -usertypes option to determine how JPublisher will implement the custom Java class that corresponds to a Oracle object type:

A setting of -usertypes=oracle, which is the default setting, instructs JPublisher to create a ORAData implementation for the custom object class. This will also result in JPublisher producing a ORAData implementation for the corresponding custom reference class. A setting of -usertypes=jdbc instructs JPublisher to create a SQLData implementation for the custom object class. No custom reference class can be created. You must use java.sql.Ref or oracle.sql.REF for the reference type.
Note: You can also use JPublisher with a -usertypes=oracle setting in creating ORAData implementations to map SQL collection types.

The -usertypes=jdbc setting is not valid for mapping SQL collection types. The SQLData interface is intended only for mapping Oracle object types. Mapping Attribute Types to Java If you do not specify mappings for the attribute types of the Oracle object type, then JPublisher uses the following defaults:

For numeric attribute types, the default mapping is object-JDBC. For LOB attribute types, the default mapping is Oracle. For built-in type attribute types, the default mapping is JDBC.

If you want alternate mappings, then use the -numbertypes, -lobtypes, and -builtintypes options, as necessary, depending on the attribute types you have and the mappings you desire. If an attribute type is itself an Oracle object type, then it will be mapped according to the -usertypes setting.

13-32 Oracle Database JDBC Developer's Guide and Reference

Describing an Object Type

Important: Be aware that if you specify an SQLData implementation for the custom object class and want the code to be portable, then you must be sure to use portable mappings for the attribute types. The defaults for numeric types and built-in types are portable, but for LOB types you must specify -lobtypes=jdbc.

Summary of SQL Type Categories and Mapping Settings Table 131 summarizes JPublisher categories for SQL types, the mapping settings relevant for each category, and the default settings.
Table 131 SQL Type Category UDT types numeric types LOB types built-in types JPublisher SQL Type Categories, Supported Settings, and Defaults JPublisher Mapping Option -usertypes -numbertypes -lobtypes -builtintypes Mapping Settings oracle, jdbc oracle, jdbc, objectjdbc, bigdecimal oracle, jdbc oracle, jdbc Default oracle objectjdbc oracle jdbc

Describing an Object Type

Oracle JDBC includes functionality to retrieve information about a structured object type regarding its attribute names and types. This is similar conceptually to retrieving information from a result set about its column names and types, and in fact uses an almost identical method. This section covers the following topics:

Functionality for Getting Object Metadata Steps for Retrieving Object Metadata

Functionality for Getting Object Metadata

The oracle.sql.StructDescriptor class includes functionality to retrieve metadata about a structured object type. The StructDescriptor class has a getMetaData method with the same functionality as the standard getMetaData method available in result set objects. It returns a set of attribute information, such as attribute names and types. Call this method on a StructDescriptor object to get metadata about the Oracle object type that the StructDescriptor object describes. The signature of the StructDescriptor class getMetaData method is the same as the signature specified for getMetaData in the standard ResultSet interface. The signature is as follows:
ResultSetMetaData getMetaData() throws SQLException

However, this method actually returns an instance of oracle.jdbc.StructMetaData, a class that supports structured object metadata in the same way that the standard java.sql.ResultSetMetaData interface specifies support for result set metadata. The following method is also supported by StructMetaData:
String getOracleColumnClassName(int column) throws SQLException

Working with Oracle Object Types

13-33

Describing an Object Type

This method returns the fully qualified name of the oracle.sql.Datum subclass whose instances are manufactured if the OracleResultSet class getOracleObject method is called to retrieve the value of the specified attribute. For example, oracle.sql.NUMBER. To use the getOracleColumnClassName method, you must cast the ResultSetMetaData object, which that was returned by the getMetaData method, to StructMetaData.
Note:

In all the preceding method signatures, column is something of a misnomer. Where you specify a value of 4 for column, you really refer to the fourth attribute of the object.

Steps for Retrieving Object Metadata

Use the following steps to obtain metadata about a structured object type:
1. 2. 3.

Create or acquire a StructDescriptor instance that describes the relevant structured object type. Call the getMetaData method on the StructDescriptor instance. Call the metadata getter methods, getColumnName, getColumnType, and getColumnTypeName, as desired.
Note: If one of the structured object attributes is itself a structured object, repeat steps 1 through 3.

Example The following method shows how to retrieve information about the attributes of a structured object type. This includes the initial step of creating a StructDescriptor instance.
// // Print out the ADT's attribute names and types // void getAttributeInfo (Connection conn, String type_name) throws SQLException { // get the type descriptor StructDescriptor desc = StructDescriptor.createDescriptor (type_name, conn); // get type metadata ResultSetMetaData md = desc.getMetaData (); // get # of attrs of this type int numAttrs = desc.length (); // temporary buffers String attr_name; int attr_type; String attr_typeName; System.out.println ("Attributes of "+type_name+" :"); for (int i=0; i<numAttrs; i++) { attr_name = md.getColumnName (i+1); attr_type = md.getColumnType (i+1); 13-34 Oracle Database JDBC Developer's Guide and Reference

Describing an Object Type

System.out.println (" index"+(i+1)+" name="+attr_name+" type="+attr_type); // drill down nested object if (attrType == OracleTypes.STRUCT) { attr_typeName = md.getColumnTypeName (i+1); // recursive calls to print out nested object metadata getAttributeInfo (conn, attr_typeName); } } }

Working with Oracle Object Types

13-35

Describing an Object Type

13-36 Oracle Database JDBC Developer's Guide and Reference

14
Working with LOBs and BFILEs
This chapter describes how to use Java Database Connectivity (JDBC) and the oracle.sql.* classes to access and manipulate large object (LOB) and binary file (BFILE) locators and data. This chapter contains the following sections:

Oracle Extensions for LOBs and BFILEs Working with BLOBs, CLOBs and NCLOBs Data Interface for LOBs Working With Temporary LOBs Using Open and Close With LOBs Working with BFILEs Oracle SecureFiles
Notes:

In Oracle Database 11g, the Oracle JDBC drivers support the JDBC 4.0 java.sql.NClob interface. In Oracle Database 10g, the Oracle JDBC drivers support the JDBC 3.0 java.sql.Clob and java.sql.Blob interfaces. Certain Oracle extensions made in oracle.sql.CLOB and oracle.sql.BLOB in earlier Oracle Database releases are no longer necessary and are deprecated. You should port your application to the standard JDBC 3.0 interface. Prior to Oracle Database 10g, the maximum size of a LOB was 232 bytes. This restriction has been removed since Oracle Database 10g, and the maximum size is limited to the size of available physical storage. The Java LOB application programming interface (API) has not changed.

Oracle Extensions for LOBs and BFILEs

LOBs are stored in a way that optimizes space and provides efficient access. The JDBC drivers provide support for three types of LOB: binary large object (BLOB), which is used for unstructured binary data, character large object (CLOB), which is used for character data, and national character large object (NCLOB), which is used for national character data. BLOB, CLOB, and NCLOB data is accessed and referenced by using a locator that is stored in the database table and points to the BLOB, CLOB, and NCLOB data, which is outside the table.

Working with LOBs and BFILEs 14-1

Working with BLOBs, CLOBs and NCLOBs

BFILEs are large binary data objects stored in operating system files outside of database tablespaces. These files use reference semantics. They can also be located on tertiary storage devices, such as hard disks, CD-ROMs, PhotoCDs, and DVDs. As with BLOB, CLOB, and NCLOBs, a BFILE is accessed and referenced by a locator which is stored in the database table and points to the BFILE data. To work with LOB data, you must first obtain a LOB locator. Then you can read or write LOB data and perform data manipulation. The JDBC drivers support the following oracle.sql.* classes for BLOBs, CLOBs, NCLOBs, and BFILEs:

oracle.sql.BLOB oracle.sql.CLOB oracle.sql.NCLOB oracle.sql.BFILE

The oracle.sql.BLOB, oracle.sql.CLOB, and oracle.sql.NCLOB classes implement the java.sql.Blob, java.sql.Clob, and java.sql.NClob interfaces, respectively. In contrast, BFILE is an Oracle extension, without a corresponding java.sql interface. Instances of these classes contain only the locators for these data types, not the data. After accessing the locators, you must perform some additional steps to access the data.
Note: If you want to create a new LOB, then use the factory methods from oracle.jdbc.OracleConnection interface.

Working with BLOBs, CLOBs and NCLOBs

This section describes how to read and write data to and from BLOBs, CLOBs, and NCLOBs in Oracle Database, using LOB locators. This section covers the following topics:

Getting and Passing BLOB, CLOB, and NCLOB Locators Reading and Writing BLOB, CLOB and NCLOB Data Creating and Populating a BLOB, CLOB or NCLOB Column Accessing and Manipulating BLOB, CLOB, or NCLOB Data

Getting and Passing BLOB, CLOB, and NCLOB Locators

Standard as well as Oracle-specific getter and setter methods are available for retrieving or passing LOB locators from or to the database. This section covers the following topics:

Retrieving BLOB, CLOB, and NCLOB Locators Passing BLOB, CLOB and NCLOB Locators

Retrieving BLOB, CLOB, and NCLOB Locators

Given a standard JDBC result set or callable statement that includes BLOB, CLOB, or NCLOB locators, you can access the locators by using standard getter methods. You can use the standard getBlob, getClob, and getNClob methods, which return java.sql.Blob, Clob, and NClob objects, respectively.
14-2 Oracle Database JDBC Developer's Guide and Reference

Working with BLOBs, CLOBs and NCLOBs

Note:

All the standard and Oracle-specific getter methods discussed here take either an int column index or a String column name as input.

If you retrieve or cast the result set or the callable statement to OracleResultSet or OracleCallableStatement, then you can use Oracle extensions, as follows:

You can use getBLOB, getCLOB, and getNCLOB, which return oracle.sql.BLOB, CLOB, and NCLOB objects, respectively. You can also use the getOracleObject method, which returns an oracle.sql.Datum object, and cast the output appropriately.

Example: Getting BLOB, CLOB, and NCLOB Locators from a Result Set Assume the database has a table called lob_table with a column for a BLOB locator, blob_col, a column for a CLOB locator, clob_col, and a column for a NCLOB locator, nclob_col. This example assumes that you have already created the Statement object, stmt. First, select the LOB locators into a standard result set, then get the LOB data into appropriate Java classes:
// Select LOB locator into standard result set. ResultSet rs = stmt.executeQuery ("SELECT blob_col, clob_col, nclob_col FROM lob_table"); while (rs.next()) { // Get LOB locators into Java container classes. java.sql.Blob blob = (java.sql.Blob)rs.getObject(1); java.sql.Clob clob = (java.sql.Clob)rs.getObject(2); java.sql.NClob nclob = (java.sql.NClob)rs.getObject(3); (...process...) }

The output is cast to java.sql.Blob, java.sql.Clob, and java.sql.NClob. As an alternative, you can cast the output to oracle.sql.BLOB, oracle.sql.CLOB, and oracle.sql.NCLOB to take advantage of extended functionality offered by the oracle.sql.* classes. For example, you can rewrite the preceding code to get the LOB locators as:
// Get LOB locators into Java container classes. oracle.sql.BLOB blob = (BLOB)rs.getObject(1); oracle.sql.CLOB clob = (CLOB)rs.getObject(2); oracle.sql.NCLOB nclob = (NCLOB)rs.getObject(3); (...process...)

Example: Getting a CLOB Locator from a Callable Statement The callable statement methods for retrieving LOBs are identical to the result set methods. For example, if you have an OracleCallableStatement instance, ocs, that calls a function func that has a CLOB output parameter, then set up the callable statement as in the following example. This example registers OracleTypes.CLOB as the type code of the output parameter.
OracleCallableStatement ocs = (OracleCallableStatement)conn.prepareCall("{? = call func()}"); Working with LOBs and BFILEs 14-3

Working with BLOBs, CLOBs and NCLOBs

ocs.registerOutParameter(1, OracleTypes.CLOB); ocs.execute(); oracle.sql.CLOB clob = ocs.getCLOB(1);

Passing BLOB, CLOB and NCLOB Locators

Given a standard JDBC prepared statement or callable statement, you can use standard setter methods to pass LOB locators. These methods are defined as follows:
public void setBlob(int index, Blob value); public void setClob(int index, Clob value); public void setNClob(int index, NClob value);

Note:

If you pass a BLOB to a PL/SQL procedure, then the BLOB must be no bigger than 32K - 7. If you pass a BLOB that exceeds this limit, then you will receive a SQLException exception.

Given an Oracle-specific OraclePreparedStatement or OracleCallableStatement, then you can use Oracle extensions as follows:

Use setBLOB, setCLOB, and setNClob, which take oracle.sql.BLOB, CLOB and NCLOB locators as input, respectively. Use the setOracleObject method, which simply specifies an oracle.sql.Datum input.

Example: Passing a BLOB Locator to a Prepared Statement If you have an OraclePreparedStatement object ops and a BLOB named my_blob, then write the BLOB to the database as follows:
OraclePreparedStatement ops = (OraclePreparedStatement)conn.prepareStatement ("INSERT INTO blob_table VALUES(?)"); ops.setBLOB(1, my_blob); ops.execute();

Example: Passing a CLOB Locator to a Callable Statement If you have an OracleCallableStatement object ocs and a CLOB named my_clob, then input the CLOB to the stored procedure proc as follows:
OracleCallableStatement ocs = (OracleCallableStatement)conn.prepareCall("{call proc(?))}"); ocs.setClob(1, my_clob); ocs.execute();

Example: Passing an NCLOB Locator to a Callable Statement If you have an OracleCallableStatement object ocs and an NCLOB named my_nclob, then input the NCLOB to the stored procedure proc as follows:
OracleCallableStatement ocs = (OracleCallableStatement)conn.prepareCall("{call proc(?))}"); ocs.setNClob(1, my_nclob); ocs.execute();

14-4 Oracle Database JDBC Developer's Guide and Reference

Working with BLOBs, CLOBs and NCLOBs

Reading and Writing BLOB, CLOB and NCLOB Data

Once you have a LOB locator, you can use JDBC methods to read and write the LOB data. LOB data is materialized as a Java array or stream. Unlike LONG and LONG RAW data, you can access the LOB data at any time during the life of the connection. To read and write the LOB data, use the methods in the java.sql.BLOB, java.sql.CLOB, and java.sql.NCLOB class, as appropriate. These classes provide functionality such as reading from the LOB into an input stream, writing from an output stream into a LOB, determining the length of a LOB, and closing a LOB.
Notes: To write LOB data, the application must acquire a write lock on the LOB object. One way to accomplish this is through a SELECT FOR UPDATE. Also, you must disable auto-commit mode.

To read and write LOB data, you can use these methods:

To read from a BLOB, use the getBinaryStream method of an java.sql.BLOB object to retrieve the entire BLOB as an input stream. This returns a java.io.InputStream object. As with any InputStream object, use one of the overloaded read methods to read the LOB data and use the close method when you finish.

To write to a BLOB, use the setBinaryStream method of an java.sql.BLOB object to retrieve the BLOB as an output stream. This returns a java.io.OutputStream object to be written back to the BLOB. As with any OutputStream object, use one of the overloaded write methods to update the LOB data and use the close method when you finish.

To read from a CLOB, use the getAsciiStream or getCharacterStream method of an java.sql.CLOB object to retrieve the entire CLOB as an input stream. The getAsciiStream method returns an ASCII input stream in a java.io.InputStream object. The getCharacterStream method returns a Unicode input stream in a java.io.Reader object. As with any InputStream or Reader object, use one of the overloaded read methods to read the LOB data and use the close method when you finish. You can also use the getSubString method of java.sql.CLOB object to retrieve a subset of the CLOB as a character string of type java.lang.String.

To write to a CLOB, use the setAsciiStream or setCharacterStream method of an java.sql.CLOB object to retrieve the CLOB as an output stream to be written back to the CLOB. The setAsciiStream method returns an ASCII output stream in a java.io.OutputStream object. The setCharacterStream method returns a Unicode output stream in a java.io.Writer object. As with any Stream or Writer object, use one of the overloaded write methods to update the LOB data and use the flush and close methods when you finish.

To read from an NCLOB, use the getAsciiStream or getCharacterStream method of an java.sql.NCLOB object to retrieve the entire NCLOB as an input stream. The getAsciiStream method returns an ASCII input stream in a java.io.InputStream object. The getCharacterStream method returns a Unicode input stream in a java.io.Reader object. As with any InputStream or Reader object, use one of the overloaded read methods to read the LOB data and use the close method when you finish.

Working with LOBs and BFILEs 14-5

Working with BLOBs, CLOBs and NCLOBs

You can also use the getSubString method of java.sql.NCLOB object to retrieve a subset of the NCLOB as a character string of type java.lang.String.

To write to an NCLOB, use the setAsciiStream or setCharacterStream method of an oracle.sql.NCLOB object to retrieve the NCLOB as an output stream to be written back to the NCLOB. The setAsciiStream method returns an ASCII output stream in a java.io.OutputStream object. The setCharacterStream method returns a Unicode output stream in a java.io.Writer object. As with any Stream or Writer object, use one of the overloaded write methods to update the LOB data and use the flush and close methods when you finish.
Notes:

The stream write methods described in this section write directly to the database when you write to the output stream. You do not need to run an UPDATE to write the data. However, you need to call close or flush to ensure all changes are written. CLOBs and BLOBs are transaction controlled. After writing to either, you must commit the transaction for the changes to be permanent. BFILEs are not transaction controlled. Once you write to them the changes are permanent, even if the transaction is rolled back, unless the external file system does something else. When writing to or reading from a CLOB or an NCLOB, the JDBC drivers perform all character set conversions for you. When reading a LOB using any of the getXXX methods described in the section, the returned stream fetches blocks of data from the database needed. The entire LOB is not fetched all at once, which makes it practical to read very large LOBs.

Example: Reading BLOB Data Use the getBinaryStream method of the oracle.sql.BLOB class to read BLOB data. The getBinaryStream method provides access to the BLOB data through a binary stream. The following example uses the getBinaryStream method to read BLOB data through a byte stream and then reads the byte stream into a byte array, returning the number of bytes read, as well.
// Read BLOB data from BLOB locator. InputStream byte_stream = my_blob.getBinaryStream(1L); byte [] byte_array = new byte [10]; int bytes_read = byte_stream.read(byte_array); ...

Example: Reading CLOB Data The following example uses the getCharacterStream method to read CLOB data into a Unicode character stream. It then reads the character stream into a character array, returning the number of characters read, as well.
// Read CLOB data from CLOB locator into Reader char stream. Reader char_stream = my_clob.getCharacterStream(1L); char [] char_array = new char [10]; int chars_read = char_stream.read (char_array, 0, 10);

14-6 Oracle Database JDBC Developer's Guide and Reference

Working with BLOBs, CLOBs and NCLOBs

...

Example: Reading NCLOB Data The following example uses the getCharacterStream method to read NCLOB data into a Unicode character stream. It then reads the character stream into a character array, returning the number of characters read, as well.
// Read NCLOB data from NCLOB locator into Reader char stream. Reader char_stream = my_nclob.getCharacterStream(1L); char [] char_array = new char [10]; int chars_read = char_stream.read (char_array, 0, 10); ...

The next example uses the getAsciiStream method of the oracle.sql.NCLOB class to read NCLOB data through an ASCII character stream. It then reads the ASCII stream into a byte array, returning the number of bytes read, as well.
// Read NCLOB data from NCLOB locator into Input ASCII character stream Inputstream asciiChar_stream = my_nclob.getAsciiStream(1L); byte[] asciiChar_array = new byte[10]; int asciiChar_read = asciiChar_stream.read(asciiChar_array,0,10);

Example: Writing BLOB Data Use the setBinaryOutputStream method of an oracle.sql.BLOB object to write BLOB data. The following example reads a vector of data into a byte array, then uses the setBinaryOutputStream method to write an array of character data to a BLOB.
java.io.OutputStream outstream; // read data into a byte array byte[] data = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}; // write the array of binary data to a Blob outstream = ((BLOB)my_blob).setBinaryOutputStream(1L); outstream.write(data); ...

Example: Writing CLOB Data Use the setCharacterStream method or the setAsciiStream method to write data to a CLOB. The setCharacterStream method returns a Unicode output stream. The setAsciiStream method returns an ASCII output stream. The following example reads a vector of data into a character array, then uses the setCharacterStream method to write the array of character data to a CLOB.
java.io.Writer writer; // read data into a character array char[] data = {'0','1','2','3','4','5','6','7','8','9'}; // write the array of character data to a Clob writer = ((CLOB)my_clob).setCharacterStream(); writer.write(data); writer.flush(); writer.close(); ...

Working with LOBs and BFILEs 14-7

Working with BLOBs, CLOBs and NCLOBs

The next example reads a vector of data into a byte array, then uses the setAsciiStream method to write the array of ASCII data to a CLOB.
java.io.OutputStream out; // read data into a byte array byte[] data = {'0','1','2','3','4','5','6','7','8','9'}; // write the array of ascii data to a CLOB out = clob.setAsciiStream(); out.write(data); out.flush(); out.close();

Example: Writing NCLOB Data Use the setCharacterStream method or the setAsciiStream method to write data to an NCLOB. The setCharacterStream method returns a Unicode output stream. The setAsciiStream method returns an ASCII output stream. The following example reads a vector of data into a character array, then uses the setCharacterStream method to write the array of character data to an NCLOB.
java.io.Writer writer; // read data into a character array char[] data = {'0','1','2','3','4','5','6','7','8','9'}; // write the array of character data to an NClob writer = ((NCLOB)my_nclob).setCharacterStream(); writer.write(data); writer.flush(); writer.close(); ...

The next example reads a vector of data into a byte array, then uses the setAsciiStream method to write the array of ASCII data to an NCLOB.
java.io.OutputStream out; // read data into a byte array byte[] data = {'0','1','2','3','4','5','6','7','8','9'}; // write the array of ascii data to a NClob out = nclob.setAsciiStream(); out.write(data); out.flush(); out.close();

Creating and Populating a BLOB, CLOB or NCLOB Column

Create and populate aBLOB, CLOB, or NCLOB column in a table by using SQL statements. You cannot construct a new BLOB, CLOB, or NCLOB locator in your application with a Java new statement. You must create the locator through a SQL operation, and then select it into your application or use the factory methods from oracle.jdbc.OracleConnection interface.
Note:

14-8 Oracle Database JDBC Developer's Guide and Reference

Working with BLOBs, CLOBs and NCLOBs

Create a BLOB, CLOB, or NCLOB column in a table with the SQL CREATE TABLE statement, then populate the LOB. This includes creating the LOB entry in the table, obtaining the LOB locator, and then copying the data into the LOB. Creating a BLOB, CLOB, or NCLOB Column in a New Table To create a BLOB, CLOB, or NCLOB column in a new table, run the SQL CREATE TABLE statement. The following example code creates a BLOB column in a new table. This example assumes that you have already created your Connection object conn and Statement object stmt:
String cmd = "CREATE TABLE my_blob_table (x VARCHAR2 (30), c BLOB)"; stmt.execute (cmd);

In this example, the VARCHAR2 column designates a row number, such as 1 or 2, and the BLOB column stores the locator of the BLOB data. Populating a BLOB, CLOB, or NCLOB Column in a New Table This example demonstrates how to populate a BLOB, CLOB, or NCLOB column by reading data from a stream. These steps assume that you have already created your Connection object conn and Statement object stmt. The table my_blob_table is the table that was created in the previous section. The following example writes the john.gif file to a BLOB:
1.

Begin by using SQL statements to create the BLOB entry in the table. Use the empty_blob function to create the BLOB locator.
stmt.execute ("INSERT INTO my_blob_table VALUES ('row1', empty_blob())");

2.

Get the BLOB locator from the table.

BLOB blob; cmd = "SELECT * FROM my_blob_table WHERE X='row1' FOR UPDATE"; ResultSet rset = stmt.executeQuery(cmd); rset.next(); BLOB blob = ((OracleResultSet)rset).getBLOB(2);

Note:
3.

You must disable auto-commit mode.

Declare a file handler for the john.gif file, then print the length of the file. This value will be used later to ensure that the entire file is read into the BLOB. Next, create a FileInputStream object to read the contents of the file, and an OutputStream object to retrieve the BLOB as a stream.
File binaryFile = new File("john.gif"); System.out.println("john.gif length = " + binaryFile.length()); FileInputStream instream = new FileInputStream(binaryFile); OutputStream outstream = blob.setBinaryStream(1L);

4.

Call getBufferSize to retrieve the ideal buffer size to use in writing to the BLOB, then create the buffer byte array.
int size = blob.getBufferSize(); byte[] buffer = new byte[size]; int length = -1;

Working with LOBs and BFILEs 14-9

Data Interface for LOBs

5.

Use the read method to read the file to the byte array buffer, then use the write method to write it to the BLOB. When you finish, close the input and output streams and commit the changes.
while ((length = instream.read(buffer)) != -1) outstream.write(buffer, 0, length); instream.close(); outstream.close(); conn.commit();

Once your data is in the BLOB, CLOB, or NCLOB, you can manipulate the data.

Accessing and Manipulating BLOB, CLOB, or NCLOB Data

Once you have your BLOB, CLOB, or NCLOB locator in a table, you can access and manipulate the data to which it points. To access and manipulate the data, you first must select their locators from a result set or from a callable statement. After you select the locators, you can retrieve the BLOB, CLOB, or NCLOB data. After retrieving the BLOB, CLOB, or NCLOB data, you can manipulate it however you want. This example is a continuation of the example in the previous section. It uses the SQL SELECT statement to select the BLOB locator from the table my_blob_table into a result set. The result of the data manipulation is to print the length of the BLOB in bytes.
// Select the blob - what we are really doing here // is getting the blob locator into a result set BLOB blob; cmd = "SELECT * FROM my_blob_table"; ResultSet rset = stmt.executeQuery (cmd); // Get the blob data - cast to OracleResult set to // retrieve the data in oracle.sql format String index = ((OracleResultSet)rset).getString(1); blob = ((OracleResultSet)rset).getBLOB(2); // get the length of the blob int length = blob.length(); // print the length of the blob System.out.println("blob length" + length); // read the blob into a byte array // then print the blob from the array byte bytes[] = blob.getBytes(1, length); blob.printBytes(bytes, length);

Data Interface for LOBs

The data interface for LOBs provides a streamlined mechanism for writing and reading the entire LOB contents. It is simpler to code and faster in many cases. It does not provide the random access capability or access beyond 2147483648 elements as do the standard java.sql.Blob and java.sql.Clob interfaces and the Oracle extensions, oracle.sql.BLOB, oracle.sql.BFILE, and oracle.sql.CLOB.

14-10 Oracle Database JDBC Developer's Guide and Reference

Data Interface for LOBs

Input In Oracle Database 10g release 2 (10.2), the setBytes, setBinaryStream, setString, setCharacterStream, and setAsciiStream methods of PreparedStatement are extended for BLOB and CLOB parameters. For the JDBC Oracle Call Interface (OCI) and Thin drivers there is no limitation on the size of the byte array or String and no limit on the length specified for the stream functions except the limits imposed by the Java language, which is that array sizes are limited to positive Java int or 2147483648 elements. For the server-side internal driver there is currently a limitation of 4000 bytes for operations on SQL statements, such as an INSERT statement. The limitation does not apply for PL/SQL statements. There is a simple workaround for an INSERT statement, which is to wrap it in a PL/SQL block, as follows:
BEGIN INSERT id, c INTO clob_tab VALUES(?,?); END;

You must bear in mind the following automatic switching of the input mode for large data:

For SQL statements: setBytes switches to setBinaryStream for data larger than 2000 bytes setString switches to setCharacterStream for data larger than 32766 characters

PL/SQL statements setBytes switches to setBinaryStream for data larger than 2000 bytes and to setBytesForBlob for data larger that 32766 bytes setString switches to setStringForClob for string data larger than 32766 bytes in the database character set setNClob or setObject is used for large national character set type data. If the setObject method is used , the target data type must be specified as Types.NCHAR, Types.NCLOB, Types.NVARCHAR, or Types.LONGNVARCHAR.

This will have impact on some programs, which formerly got ORA-17157 errors for attempts to use setString for String values larger than 32766 characters. Now, depending on the type of the target parameter an error may occur while the application is executed or the operation may succeed. Another impact is that the automatic switching may result in additional server-side parsing to adapt to the change in the parameter type. This would result in a performance effect if the data sizes vary above and below the limit for repeated executions of the statement. Switching to the stream modes will effect batching as well. Oracle Database 10g release 1 (10.1) has the SetBigStringTryClob connection property. Setting this property causes the standard setString method to switch to setStringForClob method for large data. This property is no longer used or needed. The setBytesForBlob and setStringForClob methods create temporary LOBs, which are automatically freed when the statement is executed or closed before execution. However, when a PL/SQL procedure or function is embedded in a SQL statement, data less than 4 KB is bound as String, which is the standard. When data is greater

Working with LOBs and BFILEs

14-11

Data Interface for LOBs

than 4KB, the driver binds the data as a String as for any SQL statement. This will throw an error. The workaround is to use setClob or setCharacterStream instead of setString or setStringForClob. You can also create a callable statement. Output The getBytes, getBinaryStream, getSting, getCharacterStream, and getAsciiStream methods of ResultSet and CallableStatement are extended to work with BLOB, CLOB, and BFILE columns or OUT parameters. These methods will work for any LOB of length less than 2147483648. This operates entirely on the client-side and will work with any supported version of the database, that is, Oracle Database 8.1.7 and later. BLOB, BFILE, or CLOB data can be read and written using the same streaming mechanism as for LONG RAW and LONG data. To read, use defineColumnType(nn, Types.LONGVARBINARY) or defineColumnType(nn,Types.LONGVARCHAR) on the column. This produces a direct stream on the data as if it were a LONG RAW or LONG column. This technique is limited to Oracle Database 10g release 1 (10.1) and later. CallableSatement and IN OUT Parameter It is a PL/SQL requirement that the Java types used as input and output for an IN OUT parameter must be the same. The automatic switching of types done by the extensions described in this chapter may cause problems with this. Consider that you have an IN OUT CLOB parameter of a stored procedure and you wish to use setString for setting the value for this parameter. For any IN and OUT parameter, the binds must be of the same type. The automatic switching of the input mode will cause problems unless you are sure of the data sizes. For example, if it is known that neither the input nor output data will ever be larger than 32766 bytes, then you could use setString for the input parameter and register the OUT parameter as Types.VARCHAR and use getString for the output parameter. A better solution is to change the stored procedure to have separate IN and OUT parameters. That is, if you have:
CREATE PROCEDURE clob_proc( c IN OUT CLOB );

then, change it to:

CREATE PROCEDURE clob_proc( c_in IN CLOB, c_out OUT CLOB );

Another workaround is to use a container block to make the call. The clob_proc procedure can be wrapped with a Java string to use for the prepareCall statement, as follows:
"DECLARE c_temp; BEGIN c_temp := ?; clob_proc( c_temp); ? := c_temp; END;"

In either case you may use setString on the first parameter and registerOutParameter with Types.CLOB on the second. Size Limitations Please be aware of the effect on the performance of the Java memory management system due to creation of very large byte array or String. Please read the information provided by your Java virtual machine (JVM) vendor about the impact of very large data elements on memory management, and consider using the stream interfaces instead.

14-12 Oracle Database JDBC Developer's Guide and Reference

Working With Temporary LOBs

Working With Temporary LOBs

You can use temporary LOBs to store transient data. The data is stored in temporary table space rather than regular table space. You should free temporary LOBs after you no longer need them. If you do not, then the space the LOB consumes in temporary table space will not be reclaimed. You can insert temporary LOBs into a table. When you do this, a permanent copy of the LOB is created and stored. Inserting a temporary LOB may be preferable for some situations. For example, if the LOB data is relatively small so that the overhead of copying the data is less than the cost of a database round trip to retrieve the empty locator. Remember that the data is initially stored in the temporary table space on the server and then moved into permanent storage. You create a temporary LOB with the static method createTemporary(Connection, boolean, int). This method is defined in both the oracle.sql.BLOB and oracle.sql.CLOB classes. You free a temporary LOB with the freeTemporary method.
public static BLOB createTemporary(Connection conn, boolean isCached, int duration); public static CLOB createTemporary(Connection conn, boolean isCached, int duration);

The duration must be either DURATION_SESSION or DURATION_CALL as defined in the oracle.sql.BLOB or oracle.sql.CLOB class. In client applications, DURATION_SESSION is appropriate. In Java stored procedures, you can use either DURATION_SESSION or DURATION_CALL, which ever is appropriate. You can test whether a LOB is temporary by calling the isTemporary method. If the LOB was created by calling the createTemporary method, then the isTemporary method returns true, else it returns false. You can free a temporary LOB by calling the freeTemporary method. Free any temporary LOBs before ending the session or call. Otherwise, the storage used by the temporary LOB will not be reclaimed.
Notes:

Failure to free a temporary LOB will result in the storage used by that LOB in the database being unavailable. Frequent failure to free temporary LOBs will result in filling up temporary table space with unavailable LOB storage. When fetching data from a ReultSet with columns that are temporary LOBs, use getClob or getBlob instead of getString or getBytes. Also invoke freeTemporary to free the temporary LOBs.

Creating Temporary NCLOBs in JDK 1.5 You create temporary national character large objects (NCLOBs) using a variant of the createTemporary method.
CLOB.createTemporary (Connection conn, boolean cache, int duration, short form);

The form argument specifies whether the created LOB is a CLOB or an NCLOB. If form equals oracle.jdbc.OraclePreparedStatement.FORM_NCHAR, then the method creates an NCLOB. If form equals

Working with LOBs and BFILEs

14-13

Using Open and Close With LOBs

oracle.jdbc.OraclePreparedStatement.FORM_CHAR, then the method creates a CLOB. Creating Temporary NCLOBs in JDK 1.6 JDBC 4.0 supports NCLOBs directly. You can use the standard factory method of java.sql.Connection interface to create an NCLOB.

Using Open and Close With LOBs

You do not have to open and close your LOBs. You may choose to open and close them for performance reasons. If you do not wrap LOB operations inside an Open/Close call operation, then each modification to the LOB will implicitly open and close the LOB, thereby firing any triggers on a domain index. Note that in this case, any domain indexes on the LOB will become updated as soon as LOB modifications are made. Therefore, domain LOB indexes are always valid and may be used at any time. If you wrap your LOB operations inside the Open/Close call operation, then triggers will not be fired for each LOB modification. Instead, the trigger on domain indexes will be fired at the Close call. For example, you might design your application so that domain indexes are not be updated until you call the close method. However, this means that any domain indexes on the LOB will not be valid in-between the Open/Close calls. You open a LOB by calling the open or open(int) method. You may then read and write the LOB without any triggers associated with that LOB firing. When you are done accessing the LOB, close the LOB by calling the close method. When you close the LOB, any triggers associated with the LOB will fire. You can see if a LOB is open or closed by calling the isOpen method. If you open the LOB by calling the open(int) method, the value of the argument must be either MODE_READONLY or MODE_READWRITE, as defined in the oracle.sql.BLOB and oracle.sql.CLOB classes. If you open the LOB with MODE_READONLY, any attempt to write to the LOB will result in a SQL exception.
Note:

An error occurs if you commit the transaction before closing all LOBs that were opened by the transaction. The openness of the open LOBs is discarded, but the transaction is successfully committed. Hence, all the changes made to the LOB and non-LOB data in the transaction are committed, but the triggers for domain indexing are not fixed.

Working with BFILEs

This section describes how to read data to BFILEs, using file locators. This section covers the following topics:

Getting and Passing BFILE Locators Reading BFILE Data Creating and Populating a BFILE Column Accessing and Manipulating BFILE Data

14-14 Oracle Database JDBC Developer's Guide and Reference

Working with BFILEs

Getting and Passing BFILE Locators

Getter and setter methods are available for retrieving or passing BFILE locators from or to the database. Retrieving BFILE Locators Given a standard JDBC result set or callable statement object that includes BFILE locators, you can access the locators by using the standard result set getObject method. This method returns an oracle.sql.BFILE object. You can also access the locators by casting your result set to OracleResultSet or your callable statement to OracleCallableStatement and using the getOracleObject or getBFILE method.
Notes:

In the OracleResultSet and OracleCallableStatement classes, getBFILE and getBfile both return oracle.sql.BFILE. There is no java.sql interface for BFILEs. If using getObject or getOracleObject, remember to cast the output, as necessary.

Example: Getting a BFILE locator from a Result Set Assume that the database has a table called bfile_table with a single column for the BFILE locator bfile_col. This example assumes that you have already created your Statement object stmt. Select the BFILE locator into a standard result set. If you cast the result set to OracleResultSet, then you can use getBFILE to get the BFILE locator, as follows:
// Select the BFILE locator into a result set ResultSet rs = stmt.executeQuery("SELECT bfile_col FROM bfile_table"); while (rs.next()) { oracle.sql.BFILE my_bfile = ((OracleResultSet)rs).getBFILE(1); }

Note that as an alternative, you can use getObject to return the BFILE locator. In this case, because getObject returns a java.lang.Object, cast the results to BFILE. For example:
oracle.sql.BFILE my_bfile = (BFILE)rs.getObject(1);

Example: Getting a BFILE Locator from a Callable Statement Assume you have an OracleCallableStatement object ocs that calls a function func that has a BFILE output parameter. The following code example sets up the callable statement, registers the output parameter as OracleTypes.BFILE, runs the statement, and retrieves the BFILE locator:
OracleCallableStatement ocs = (OracleCallableStatement)conn.prepareCall("{? = call func()}"); ocs.registerOutParameter(1, OracleTypes.BFILE); ocs.execute(); oracle.sql.BFILE bfile = ocs.getBFILE(1);

Working with LOBs and BFILEs

14-15

Working with BFILEs

Passing BFILE Locators To pass a BFILE locator to a prepared statement or callable statement, you can do one of the following:

Use the standard setObject method. Cast the statement to OraclePreparedStatement or OracleCallableStatement, and use the setOracleObject or setBFILE method.

These methods take the parameter index and an oracle.sql.BFILE object as input. Example: Passing a BFILE Locator to a Prepared Statement Assume you want to insert a BFILE locator into a table, and you have an OraclePreparedStatement object ops to insert data into a table. The first column is a string, the second column is a BFILE, and you have a valid oracle.sql.BFILE object, bfile. Write the BFILE to the database, as follows:
OraclePreparedStatement ops = (OraclePreparedStatement)conn.prepareStatement ("INSERT INTO my_bfile_table VALUES (?,?)"); ops.setString(1,"one"); ops.setBFILE(2, bfile); ops.execute();

Example: Passing a BFILE Locator to a Callable Statement Passing a BFILE locator to a callable statement is similar to passing it to a prepared statement. In this case, the BFILE locator is passed to the myGetFileLength procedure, which returns the BFILE length as a numeric value.
OracleCallableStatement cstmt = (OracleCallableStatement)conn.prepareCall ("begin ? := myGetFileLength (?); end;"); try { cstmt.registerOutParameter (1, Types.NUMERIC); cstmt.setBFILE (2, bfile); cstmt.execute (); return cstmt.getLong (1); }

Reading BFILE Data

To read BFILE data, you must first get the BFILE locator. You can get the locator from either a callable statement or a result set. Once you obtain the locator, you can call a number of methods on the BFILE without opening it. For example, you can use the oracle.sql.BFILE methods fileExists() and isFileOpen() to determine whether the BFILE exists and if it is open. However, if you want to read and manipulate the data, then you must open and close the BFILE, as follows:

Use the openFile method of the oracle.sql.BFILE class to open a BFILE. When you are done, use the closeFile method of the BFILE class.

BFILE data is through a Java stream. To read from a BFILE, use the getBinaryStream method of an oracle.sql.BFILE object to access the file as an input stream. This returns a java.io.InputStream object. As with any InputStream object, use one of the overloaded read methods to read the file data and use the close method when you finish.

14-16 Oracle Database JDBC Developer's Guide and Reference

Working with BFILEs

Notes:

BFILEs are read-only. They reside in the OS (operating system) File System and can be written to only using OS tools and commands. You can create a BFILE. However, you cannot create an OS file that a BFILE would refer to. Those are created only externally.

Example: Reading BFILE Data The following example uses the getBinaryStream method of an oracle.sql.BFILE object to read BFILE data into a byte stream and then read the byte stream into a byte array. The example assumes that the BFILE has already been opened.
// Read BFILE data from a BFILE locator Inputstream in = bfile.getBinaryStream(); byte[] byte_array = new byte{10}; int byte_read = in.read(byte_array);

Creating and Populating a BFILE Column

This section discusses how to create a BFILE column in a table with SQL operations and specify the location where the BFILE resides. The examples in this section assume that you have already created your Connection object conn and Statement object stmt. Creating a BFILE Column in a New Table To work with BFILE data, create a BFILE column in a table, and specify the location of the BFILE. To specify the location of the BFILE, use the SQL CREATE DIRECTORY...AS statement to specify an alias for the directory where the BFILE resides. In this example, the directory alias is test_dir and the BFILE resides in the /home/work directory.
String cmd; cmd = "CREATE DIRECTORY test_dir AS '/home/work'"; stmt.execute (cmd);

Use the SQL CREATE TABLE statement to create a table containing a BFILE column. In this example, the name of the table is my_bfile_table.
// Create a table containing a BFILE field cmd = "CREATE TABLE my_bfile_table (x varchar2 (30), b bfile)"; stmt.execute (cmd);

In this example, the VARCHAR2 column designates a row number and the BFILE column stores the locator of the BFILE data. Populating a BFILE Column Use the SQL INSERT INTO...VALUES statement to populate the VARCHAR2 and BFILE fields. The BFILE column is populated with the locator to the BFILE data. To populate the BFILE column, use the bfilename function to specify the directory alias and the name of the BFILE file.
cmd ="INSERT INTO my_bfile_table VALUES ('one', bfilename(test_dir, 'file1.data'))"; stmt.execute (cmd); cmd ="INSERT INTO my_bfile_table VALUES ('two', bfilename(test_dir,

Working with LOBs and BFILEs

14-17

Working with BFILEs

'jdbcTest.data'))"; stmt.execute (cmd);

In this example, the name of the directory alias is test_dir. The locator of the BFILE file1.data is loaded into the BFILE column on row one, and the locator of the BFILE jdbcTest.data is loaded into the bfile column on row two. As an alternative, you may want to create the row for the row number and BFILE locator now, but wait until later to insert the locator. In this case, insert the row number into the table and null as a place holder for the BFILE locator.
cmd ="INSERT INTO my_bfile_table VALUES ('three', null)"; stmt.execute(cmd);

Here, three is inserted into the row number column and null is inserted as the place holder. Later in your program, insert the BFILE locator into the table by using a prepared statement. First get a valid BFILE locator into the bfile object:
rs = stmt.executeQuery("SELECT b FROM my_bfile_table WHERE x='two'"); rs.next(); oracle.sql.BFILE bfile = ((OracleResultSet)rs).getBFILE(1);

Then, create your prepared statement. Note that because this example uses the setBFILE method to identify the BFILE, the prepared statement must be cast to OraclePreparedStatement:
OraclePreparedStatement ops = (OraclePreparedStatement)conn.prepareStatement (UPDATE my_bfile_table SET b=? WHERE x = 'three'); ops.setBFILE(1, bfile); ops.execute();

Now row two and row three contain the same BFILE. Once you have the BFILE locators available in a table, you can access and manipulate the BFILE data.

Accessing and Manipulating BFILE Data

Once you have the BFILE locator in a table, you can access and manipulate the data to which it points. To access and manipulate the data, you must first select its locator from a result set or a callable statement. The following code continues the example from the preceding section, getting the locator of the BFILE from row two of a table into a result set. The result set is cast to OracleResultSet so that oracle.sql.* methods can be used on it. Several of the methods applied to the BFILE, such as getDirAlias and getName, do not require you to open the BFILE. Methods that manipulate the BFILE data, such as reading, getting the length, and displaying, do require you to open the BFILE. When you finish manipulating the BFILE data, you must close the BFILE.
// select the bfile locator cmd = "SELECT * FROM my_bfile_table WHERE x = 'two'"; rset = stmt.executeQuery (cmd); if (rset.next ()) BFILE bfile = ((OracleResultSet)rset).getBFILE (2); // for these methods, you do not have to open the bfile println("getDirAlias() = " + bfile.getDirAlias());

14-18 Oracle Database JDBC Developer's Guide and Reference

Oracle SecureFiles

println("getName() = " + bfile.getName()); println("fileExists() = " + bfile.fileExists()); println("isFileOpen() = " + bfile.isFileOpen()); // now open the bfile to get the data bfile.openFile(); // get the BFILE data as a binary stream InputStream in = bfile.getBinaryStream(); int length ; // read the bfile data in 6-byte chunks byte[] buf = new byte[6]; while ((length = in.read(buf)) != -1) { // append and display the bfile data in 6-byte chunks StringBuffer sb = new StringBuffer(length); for (int i=0; i<length; i++) sb.append( (char)buf[i] ); System.out.println(sb.toString()); } // we are done working with the input stream. Close it. in.close(); // we are done working with the BFILE. Close it. bfile.closeFile();

Oracle SecureFiles
In Oracle Database 11g Release 1 (11.1), Oracle SecureFiles, a completely redesigned storage for LOBs, provide the following capabilities:

SecureFile compression enables users to compress data to save disk space. SecureFile encryption introduces a new encryption facility that allows for random reads and writes of the encrypted data. Deduplication enables Oracle database to automatically detect duplicate LOB data and conserve space by storing only one copy of data. LOB data path optimization includes logical cache above storage layer, read prefetching, new caching modes, and vectored IO. High performance space management.

These features are implemented in the database and are transparenly available to JDBC programs through the existing APIs. The new setLobOptions and getLobOptions APIs are described in the PL/SQL Packages and Types Reference, and may be accessed from JDBC through callable statements.
See Also: Oracle Database SecureFiles and Large Objects Developer's Guide

Working with LOBs and BFILEs

14-19

Oracle SecureFiles

14-20 Oracle Database JDBC Developer's Guide and Reference

15
Using Oracle Object References
This chapter describes Oracle extensions to standard Java Database Connectivity (JDBC) that let you access and manipulate object references. The following topics are discussed:

Oracle Extensions for Object References Overview of Object Reference Functionality Retrieving and Passing an Object Reference Accessing and Updating Object Values Through an Object Reference Custom Reference Classes with JPublisher

Oracle Extensions for Object References

Oracle supports the use of references to database objects. Oracle JDBC provides support for object references as:

Columns in a SELECT clause IN or OUT bind variables Attributes in an Oracle object Elements in a collection type object

In SQL, an object reference (REF) is strongly typed. For example, a reference to an EMPLOYEE object would be defined as an EMPLOYEE REF, not just a REF. When you select an object reference in Oracle JDBC, be aware that you are retrieving only a pointer to an object, not the object itself. You have the choice of materializing the reference as a weakly typed oracle.sql.REF instance, or a java.sql.Ref instance for portability, or materializing it as an instance of a custom Java class that you have created in advance, which is strongly typed. Custom Java classes used for object references are referred to as custom reference classes and must implement the oracle.sql.ORAData interface. The oracle.sql.REF class implements the standard java.sql.Ref interface. You can retrieve a REF instance through a result set or callable statement object, and pass an updated REF instance back to the database through a prepared statement or callable statement object. The REF class includes functionality to get and set underlying object attribute values, and get the SQL base type name of the underlying object. Custom reference classes include this same functionality, as well as having the advantage of being strongly typed. This can help you find coding errors during compilation that might not otherwise be discovered until run time.
Using Oracle Object References 15-1

Overview of Object Reference Functionality

Note:

If you are using the oracle.sql.ORAData interface for custom object classes, then you will presumably use ORAData for corresponding custom reference classes as well. However, if you are using the standard java.sql.SQLData interface for custom object classes, then you can only use weak Java types for references. The SQLData interface is for mapping SQL object types only. You can create and retrieve REF objects in your JDBC application only by running SQL statements. There is no JDBC-specific functionality for creating and retrieving REF objects. You cannot have a reference to an array, even though arrays, like objects, are structured types.

Overview of Object Reference Functionality

To access and update object data through an object reference, you must obtain the reference instance through a result set or callable statement and then pass it back as a bind variable in a prepared statement or callable statement. It is the reference instance that contains the functionality to access and update object attributes. This section covers the following topics:

Object Reference Getter and Setter Methods Key REF Class Methods

Object Reference Getter and Setter Methods

You can use the result set, callable statement, and prepared statement methods to retrieve and pass object references. Result Set and Callable Statement Getter Methods The OracleResultSet and OracleCallableStatement classes support getREF and getRef methods to retrieve REF objects as output parameters. REF objects can be retrieved either as oracle.sql.REF instances or java.sql.Ref instances. You can also use the getObject method. These methods take as input a String column name or int column index. Prepared and Callable Statement Setter Methods The OraclePreparedStatement and OracleCallableStatement classes support setREF and setRef methods to take REF objects as bind variables and pass them to the database. You can also use the setObject method. These methods take as input a String parameter name or int parameter index as well as an oracle.sql.REF instance or a java.sql.Ref instance.

Key REF Class Methods

You can use the following oracle.sql.REF class methods to retrieve the SQL object type name and retrieve and pass the underlying object data:

getBaseTypeName

15-2 Oracle Database JDBC Developer's Guide and Reference

Retrieving and Passing an Object Reference

Retrieves the fully qualified SQL structured type name of the referenced object. This is a standard method specified by the java.sql.Ref interface.

getValue Retrieves the referenced object from the database, enabling you to access its attribute values. It optionally takes a type map object. You can use the default type map of the database connection object. This method is an Oracle extension.

setValue Sets the referenced object in the database, allowing you to update its attribute values. It takes an instance of the object type, either a STRUCT instance or an instance of a custom object class, as input. This method is an Oracle extension.

Retrieving and Passing an Object Reference

This section discusses JDBC functionality for retrieving and passing object references. It covers the following topics:

Retrieving an Object Reference from a Result Set Retrieving an Object Reference from a Callable Statement Passing an Object Reference to a Prepared Statement

Retrieving an Object Reference from a Result Set

To demonstrate how to retrieve object references, the following example first defines an Oracle object type ADDRESS, which is then referenced in the PEOPLE table:
create type ADDRESS as object (street_name VARCHAR2(30), house_no NUMBER); create table PEOPLE (col1 VARCHAR2(30), col2 NUMBER, col3 REF ADDRESS);

The ADDRESS object type has two attributes: a street name and a house number. The PEOPLE table has three columns: a column for character data, a column for numeric data, and a column containing a reference to an ADDRESS object. To retrieve an object reference, follow these general steps:
1. 2. 3. 4. 5.

Use a standard SQL SELECT statement to retrieve the reference from a database table REF column. Use getREF to get the address reference from the result set into a REF object. Let Address be the Java custom class corresponding to the SQL object type ADDRESS. Add the correspondence between the Java class Address and the SQL type ADDRESS to your type map. Use the getValue method to retrieve the contents of the Address reference. Cast the output to Address.

The PEOPLE database table is defined earlier in this section. The code for the preceding steps, except the step of adding Address to the type map, is as follows:
ResultSet rs = stmt.executeQuery("SELECT col3 FROM PEOPLE");

Using Oracle Object References 15-3

Retrieving and Passing an Object Reference

while (rs.next()) { REF ref = ((OracleResultSet)rs).getREF(1); Address a = (Address) ref.getValue(); }

Note:

In the preceding code, stmt is a previously defined statement

object. As with other SQL types, you could retrieve the reference with the getObject method of your result set. Note that this would require you to cast the output. For example:
REF ref = (REF)rs.getObject(1);

There are no performance advantages in using getObject instead of getREF; however, using getREF enables you to avoid casting the output.

Retrieving an Object Reference from a Callable Statement

To retrieve an object reference as an OUT parameter in PL/SQL blocks, you must register the bind type for your OUT parameter.
1.

Cast your callable statement to OracleCallableStatement, as follows:

OracleCallableStatement ocs = (OracleCallableStatement)conn.prepareCall("{? = call func()}");

2.

Register the OUT parameter with the following form of the registerOutParameter method:
ocs.registerOutParameter (int param_index, int sql_type, String sql_type_name);

param_index is the parameter index and sql_type is the SQL type code. The sql_type_name is the name of the structured object type that this reference is used for. For example, if the OUT parameter is a reference to an ADDRESS object, then ADDRESS is the sql_type_name that should be passed in.
3.

Run the call, as follows:

ocs.execute();

Passing an Object Reference to a Prepared Statement

Pass an object reference to a prepared statement in the same way as you would pass any other SQL type. Use either the setObject method or the setREF method of a prepared statement object. Use a prepared statement to update an address reference based on ROWID, as follows:
PreparedStatement pstmt = conn.prepareStatement ("update PEOPLE set ADDR_REF = ? where ROWID = ?"); ((OraclePreparedStatement)pstmt).setREF (1, addr_ref); ((OraclePreparedStatement)pstmt).setROWID (2, rowid);

15-4 Oracle Database JDBC Developer's Guide and Reference

Custom Reference Classes with JPublisher

Accessing and Updating Object Values Through an Object Reference

You can use the REF object setValue method to update the value of an object in the database through an object reference. To do this, you must first retrieve the reference to the database object and create a Java object that corresponds to the database object. For example, you can use the code in "Retrieving and Passing an Object Reference" on page 15-3, to retrieve the reference to a database ADDRESS object, as follows:
ResultSet rs = stmt.executeQuery("SELECT col3 FROM PEOPLE"); if (rs.next()) { REF ref = rs.getREF(1); Address a = (Address)ref.getValue(); }

Then, you can create a Java Address object that corresponds to the database ADDRESS object. Use the setValue method of the REF class to set the value of the database object, as follows:
Address addr = new Address(...); ref.setValue(addr);

Here, the setValue method updates the database ADDRESS object immediately.

Custom Reference Classes with JPublisher

This chapter primarily describes the functionality of the oracle.sql.REF class, but it is also possible to access Oracle object references through custom Java classes or, more specifically, custom reference classes. Custom reference classes offer all the functionality described earlier in this chapter, as well as the advantage of being strongly typed. A custom reference class must satisfy three requirements:

It must implement the oracle.sql.ORAData interface. Note that the standard JDBC SQLData interface, which is an alternative for custom object classes, is not intended for custom reference classes. It, or a companion class, must implement the oracle.sql.ORADataFactory interface, for creating instances of the custom reference class. It must provide a way to refer to the object data. JPublisher accomplishes this by using an oracle.sql.REF attribute.

You can create custom reference classes yourself, but the most convenient way to produce them is through the Oracle JPublisher utility. If you use JPublisher to generate a custom object class to map to an Oracle object and you specify that JPublisher use a ORAData implementation, then JPublisher will also generate a custom reference class that implements ORAData and ORADataFactory and includes an oracle.sql.REF attribute. The ORAData implementation will be used if the JPublisher -usertypes mapping option is set to oracle, which is the default. Custom reference classes are strongly typed. For example, if you define an Oracle object EMPLOYEE, then JPublisher can generate an Employee custom object class and an EmployeeRef custom reference class. Using EmployeeRef instances instead of generic oracle.sql.REF instances makes it easier to catch errors during compilation instead of at run time. For example, if you accidentally assign some other kind of object reference into an EmployeeRef variable.

Using Oracle Object References 15-5

Custom Reference Classes with JPublisher

Be aware that the standard SQLData interface supports only SQL object mappings. For this reason, if you instruct JPublisher to implement the standard SQLData interface in creating a custom object class, then JPublisher will not generate a custom reference class. In this case, your only option is to use standard java.sql.Ref instances or oracle.sql.REF instances to map to your object references.
See Also:

"Using JPublisher to Create Custom Object Classes" on page 13-30 Oracle Database JPublisher User's Guide

15-6 Oracle Database JDBC Developer's Guide and Reference

16
Working with Oracle Collections
This chapter describes Oracle extensions to standard Java Database Connectivity (JDBC) that let you access and manipulate Oracle collections, which map to Java arrays, and their data. The following topics are discussed:

Oracle Extensions for Collections Overview of Collection Functionality ARRAY Performance Extension Methods Creating and Using Arrays Using a Type Map to Map Array Elements Custom Collection Classes with JPublisher

Oracle Extensions for Collections

An Oracle collection, either a variable array (VARRAY) or a nested table in the database, maps to an array in Java. JDBC 2.0 arrays are used to materialize Oracle collections in Java. The terms collection and array are sometimes used interchangeably. However, collection is more appropriate on the database side and array is more appropriate on the JDBC application side. Oracle supports only named collections, where you specify a SQL type name to describe a type of collection. JDBC enables you to use arrays as any of the following:

Columns in a SELECT clause IN or OUT bind variables Attributes in an Oracle object Elements of other arrays

This section covers the following topics:

Choices in Materializing Collections Creating Collections Creating Multilevel Collection Types

Choices in Materializing Collections

In your application, you have the choice of materializing a collection as an instance of the oracle.sql.ARRAY class, which is weakly typed, or materializing it as an instance of a custom Java class that you have created in advance, which is strongly typed. Custom Java classes used for collections are referred to as custom collection
Working with Oracle Collections 16-1

Oracle Extensions for Collections

classes. A custom collection class must implement the Oracle oracle.sql.ORAData interface. In addition, the custom class or a companion class must implement oracle.sql.ORADataFactory. The standard java.sql.SQLData interface is for mapping SQL object types only. The oracle.sql.ARRAY class implements the standard java.sql.Array interface. The ARRAY class includes functionality to retrieve the array as a whole, retrieve a subset of the array elements, and retrieve the SQL base type name of the array elements. However, you cannot write to the array, because there are no setter methods. Custom collection classes, as with the ARRAY class, enable you to retrieve all or part of the array and get the SQL base type name. They also have the advantage of being strongly typed, which can help you find coding errors during compilation that might not otherwise be discovered until run time. Furthermore, custom collection classes produced by JPublisher offer the feature of being writable, with individually accessible elements. There is no difference in the code between accessing VARRAYs and accessing nested tables. ARRAY class methods can determine if they are being applied to a VARRAY or nested table, and respond by taking the appropriate actions.
Note: See Also: For more information about custom collection classes, see "Custom Collection Classes with JPublisher" on page 16-13.

Creating Collections
Because Oracle supports only named collections, you must declare a particular VARRAY type name or nested table type name. VARRAY and nested table are not types themselves, but categories of types. A SQL type name is assigned to a collection when you create it using the SQL CREATE TYPE statement:
CREATE TYPE <sql_type_name> AS <datatype>;

A VARRAY is an array of varying size. It has an ordered set of data elements, and all the elements are of the same data type. Each element has an index, which is a number corresponding to the position of the element in the VARRAY. The number of elements in a VARRAY is the size of the VARRAY. You must specify a maximum size when you declare the VARRAY type. For example:
CREATE TYPE myNumType AS VARRAY(10) OF NUMBER;

This statement defines myNumType as a SQL type name that describes a VARRAY of NUMBER values that can contain no more than 10 elements. A nested table is an unordered set of data elements, all of the same data type. The database stores a nested table in a separate table which has a single column, and the type of that column is a built-in type or an object type. If the table is an object type, then it can also be viewed as a multi-column table, with a column for each attribute of the object type. You can create a nested table as follows:
CREATE TYPE myNumList AS TABLE OF integer;

This statement identifies myNumList as a SQL type name that defines the table type used for the nested tables of the type INTEGER.

16-2 Oracle Database JDBC Developer's Guide and Reference

ARRAY Performance Extension Methods

Creating Multilevel Collection Types

The most common way to create a new multilevel collection type in JDBC is to pass the SQL CREATE TYPE statement to the execute method of the java.sql.Statement class. The following code creates a one-level nested table, first_level, and a two- levels nested table, second_level:
Connection conn = .... // make a database // connection Statement stmt = conn.createStatement(); // open a database // cursor stmt.execute("CREATE TYPE first_level AS TABLE OF NUMBER"); // create a nested // table of number stmt.execute("CREATE second_level AS TABLE OF first_level"); // create a // two-levels nested table ... // other operations here stmt.close(); // release the // resource conn.close(); // close the // database connection

Once the multilevel collection types have been created, they can be used as both columns of a base table as well as attributes of a object type.

Overview of Collection Functionality

You can obtain collection data in an array instance through a result set or callable statement and pass it back as a bind variable in a prepared statement or callable statement. The oracle.sql.ARRAY class, which implements the standard java.sql.Array interface, provides the necessary functionality to access and update the data of an Oracle collection. This section covers Array Getter and Setter Methods. Use the following result set, callable statement, and prepared statement methods to retrieve and pass collections as Java arrays. Result Set and Callable Statement Getter Methods The OracleResultSet and OracleCallableStatement classes support getARRAY and getArray methods to retrieve ARRAY objects as output parameters, either as oracle.sql.ARRAY instances or java.sql.Array instances. You can also use the getObject method. These methods take as input a String column name or int column index. Prepared and Callable Statement Setter Methods The OraclePreparedStatement and OracleCallableStatement classes support setARRAY and setArray methods to take updated ARRAY objects as bind variables and pass them to the database. You can also use the setObject method. These methods take as input a String parameter name or int parameter index as well as an oracle.sql.ARRAY instance or a java.sql.Array instance.

ARRAY Performance Extension Methods

This section discusses the following topics:

Accessing oracle.sql.ARRAY Elements as Arrays of Java Primitive Types

Working with Oracle Collections 16-3

ARRAY Performance Extension Methods

ARRAY Automatic Element Buffering ARRAY Automatic Indexing

Accessing oracle.sql.ARRAY Elements as Arrays of Java Primitive Types

The oracle.sql.ARRAY class contains methods that return array elements as Java primitive types. These methods allow you to access collection elements more efficiently than accessing them as Datum instances and then converting each Datum instance to its Java primitive value.
Note: These specialized methods of the oracle.sql.ARRAY class are restricted to numeric collections.

Each method using the first signature returns collection elements as an XXX[], where XXX is a Java primitive type. Each method using the second signature returns a slice of the collection containing the number of elements specified by count, starting at the index location.

ARRAY Automatic Element Buffering

Oracle JDBC driver provides public methods to enable and disable buffering of ARRAY contents. The following methods are included with the oracle.sql.ARRAY class:

setAutoBuffering getAutoBuffering

It is advisable to enable auto-buffering in a JDBC application when the ARRAY elements will be accessed more than once by the getAttributes and getArray methods, presuming the ARRAY data is able to fit into the Java Virtual Machine (JVM) memory without overflow.
Important:

Buffering the converted elements may cause the JDBC application to consume a significant amount of memory.

When you enable auto-buffering, the oracle.sql.ARRAY object keeps a local copy of all the converted elements. This data is retained so that a second access of this information does not require going through the data format conversion process.

ARRAY Automatic Indexing

If an array is in auto-indexing mode, then the array object maintains an index table to hasten array element access. The oracle.sql.ARRAY class contains the following methods to support automatic array-indexing:

setAutoIndexing setAutoIndexing

By default, auto-indexing is not enabled. For a JDBC application, enable auto-indexing for ARRAY objects if random access of array elements may occur through the getArray and getResultSet methods.

16-4 Oracle Database JDBC Developer's Guide and Reference

Creating and Using Arrays

Creating and Using Arrays

This section discusses how to create array objects and how to retrieve and pass collections as array objects, including the following topics.

Creating ARRAY Objects Retrieving an Array and Its Elements Passing Arrays to Statement Objects

Creating ARRAY Objects

Note:

Oracle JDBC does not support the JDBC 4.0 method createArrayOf method of java.sql.Connection interface. This method only allows anonymous array types, while all Oracle array types are named. Use the Oracle specific method createArray of oracle.jdbc.OracleConnection instead.

This section describes how to create ARRAY objects. This section covers the following topics:

Steps in Creating ARRAY Objects Creating Multilevel Collections

Steps in Creating ARRAY Objects Starting from Oracle Database 11g Release 1 (11.1), you can use the createArray factory method of oracle.jdbc.OracleConnection interface to create an array object. The factory method for creating arrays has been defined as follows:
public ARRAY createARRAY(java.lang.String typeName,java.lang.Object elements)throws SQLException

where, typeName is the name of the SQL type of the created object and elements is the elements of the created object. Perform the following to create an array:
1.

Create a collection with the CREATE TYPE statement as follows:

CREATE TYPE elements AS varray(22) OF NUMBER(5,2);

The two possibilities for the contents of elements are:

An array of Java primitives. For example, int[]. An array of Java objects, such as xxx[], where xxx is the name of a Java class. For example, Integer[].

2.

Construct the ARRAY object by passing the Java string specifying the user-defined SQL type name of the array and a Java object containing the individual elements you want the array to contain.
ARRAY array = oracle.jdbc.OracleConnection.createARRAY(sql_type_name, elements);

Working with Oracle Collections 16-5

Creating and Using Arrays

Creating Multilevel Collections As with single-level collections, the JDBC application can create an oracle.sql.ARRAY instance to represent a multilevel collection, and then send the instance to the database. The same createArray factory method you use to create single-level collections, can be used to create multilevel collections as well. To create a single-level collection, the elements are a one dimensional Java array, while to create a multilevel collection, the elements can be either an array of oracle.sql.ARRAY[] elements or a nested Java array or the combinations. The following code shows how to create collection types with a nested Java array:
// prepare the multi level collection elements as a nested Java array int[][][] elements = { {{1}, {1, 2}}, {{2}, {2, 3}}, {{3}, {3, 4}} }; // create the ARRAY using the factory method ARRAY array = oracle.jdbc.OracleConnection.createARRAY(sql_type_name, elements);

Retrieving an Array and Its Elements

This section first discusses how to retrieve an ARRAY instance as a whole from a result set, and then how to retrieve the elements from the ARRAY instance. This section covers the following topics:

Retrieving the Array Data Retrieval Methods Comparing the Data Retrieval Methods Retrieving Elements of a Structured Object Array According to a Type Map Retrieving a Subset of Array Elements Retrieving Array Elements into an oracle.sql.Datum Array Accessing Multilevel Collection Elements

Retrieving the Array

You can retrieve a SQL array from a result set by casting the result set to OracleResultSet and using the getARRAY method, which returns an oracle.sql.ARRAY object. If you want to avoid casting the result set, then you can get the data with the standard getObject method specified by the java.sql.ResultSet interface and cast the output to oracle.sql.ARRAY.

Data Retrieval Methods

Once you have an ARRAY object, you can retrieve the data using one of these three overloaded methods of the oracle.sql.ARRAY class:

getArray getOracleArray getResultSet

Oracle also provides methods that enable you to retrieve all the elements of an array, or a subset.

16-6 Oracle Database JDBC Developer's Guide and Reference

Creating and Using Arrays

Note:

In case you are working with an array of structured objects, Oracle provides versions of these three methods that enable you to specify a type map so that you can choose how to map the objects to Java.

getOracleArray The getOracleArray method is an Oracle-specific extension that is not specified in the standard Array interface. The getOracleArray method retrieves the element values of the array into a Datum[] array. The elements are of the oracle.sql.* data type corresponding to the SQL type of the data in the original array. For an array of structured objects, this method will use oracle.sql.STRUCT instances for the elements. Oracle also provides a getOracleArray(index,count) method to get a subset of the array elements. getResultSet The getResultSet method returns a result set that contains elements of the array designated by the ARRAY object. The result set contains one row for each array element, with two columns in each row. The first column stores the index into the array for that element, and the second column stores the element value. In the case of VARRAYs, the index represents the position of the element in the array. In the case of nested tables, which are by definition unordered, the index reflects only the return order of the elements in the particular query. Oracle recommends using getResultSet when getting data from nested tables. Nested tables can have an unlimited number of elements. The ResultSet object returned by the method initially points at the first row of data. You get the contents of the nested table by using the next method and the appropriate getXXX method. In contrast, getArray returns the entire contents of the nested table at one time. The getResultSet method uses the default type map of the connection to determine the mapping between the SQL type of the Oracle object and its corresponding Java data type. If you do not want to use the default type map of the connection, another version of the method, getResultSet(map), enables you to specify an alternate type map. Oracle also provides the getResultSet(index,count) and getResultSet(index,count,map) methods to retrieve a subset of the array elements. getArray The getArray method is a standard JDBC method that returns the array elements as a java.lang.Object, which you can cast as appropriate. The elements are converted to the Java types corresponding to the SQL type of the data in the original array. Oracle also provides a getArray(index,count) method to retrieve a subset of the array elements.

Comparing the Data Retrieval Methods

If you use getOracleArray to return the array elements, then the use by that method of oracle.sql.Datum instances avoids the expense of data conversion from SQL to Java. The non-character data inside the instance of a Datum class or any of its subclass remains in raw SQL format.

Working with Oracle Collections 16-7

Creating and Using Arrays

If you use getResultSet to return an array of primitive data types, then the JDBC driver returns a ResultSet object that contains, for each element, the index into the array for the element and the element value. For example:
ResultSet rset = intArray.getResultSet();

In this case, the result set contains one row for each array element, with two columns in each row. The first column stores the index into the array and the second column stores the element value. If the elements of an array are of a SQL type that maps to a Java type, then getArray returns an array of elements of this Java type. The return type of the getArray method is java.lang.Object. Therefore, the result must be cast before it can be used.
BigDecimal[] values = (BigDecimal[]) intArray.getArray();

Here intArray is an oracle.sql.ARRAY, corresponding to a VARRAY of type NUMBER. The values array contains an array of elements of type java.math.BigDecimal, because the SQL NUMBER data type maps to Java BigDecimal, by default, according to Oracle JDBC drivers.
Note:

Using BigDecimal is a resource-intensive operation in Java. Because Oracle JDBC maps numeric SQL data to BigDecimal by default, using getArray may impact performance, and is not recommended for numeric collections.

Retrieving Elements of a Structured Object Array According to a Type Map

By default, if you are working with an array whose elements are structured objects, and you use getArray or getResultSet, then the Oracle objects in the array will be mapped to their corresponding Java data types according to the default mapping. This is because these methods use the default type map of the connection to determine the mapping. However, if you do not want default behavior, then you can use the getArray(map) or getResultSet(map) method to specify a type map that contains alternate mappings. If there are entries in the type map corresponding to the Oracle objects in the array, then each object in the array is mapped to the corresponding Java type specified in the type map. For example:
Object[] object = (Object[])objArray.getArray(map);

Where objArray is an oracle.sql.ARRAY object and map is a java.util.Map object. If the type map does not contain an entry for a particular Oracle object, then the element is returned as an oracle.sql.STRUCT object. The getResultSet(map) method behaves similarly to the getArray(map) method.
See Also:

"Using a Type Map to Map Array Elements" on page 16-12

Retrieving a Subset of Array Elements

If you do not want to retrieve the entire contents of an array, then you can use signatures of getArray, getResultSet, and getOracleArray that let you retrieve a subset. To retrieve a subset of the array, pass in an index and a count to indicate

16-8 Oracle Database JDBC Developer's Guide and Reference

Creating and Using Arrays

where in the array you want to start and how many elements you want to retrieve. As previously described, you can specify a type map or use the default type map for your connection to convert to Java types. For example:
Object object = arr.getArray(index, count, map); Object object = arr.getArray(index, count);

Similar examples using getResultSet are:

ResultSet rset = arr.getResultSet(index, count, map); ResultSet rset = arr.getResultSet(index, count);

A similar example using getOracleArray is:

Datum[] arr = arr.getOracleArray(index, count);

Where arr is an oracle.sql.ARRAY object, index is type long, count is type int, and map is a java.util.Map object.
Note: There is no performance advantage in retrieving a subset of an array, as opposed to the entire array.

Retrieving Array Elements into an oracle.sql.Datum Array

Use getOracleArray to return an oracle.sql.Datum[] array. The elements of the returned array will be of the oracle.sql.* type that correspond to the SQL data type of the elements of the original array. For example:
Datum arraydata[] = arr.getOracleArray();

arr is an oracle.sql.ARRAY object. The following example assumes that a connection object conn and a statement object stmt have already been created. In the example, an array with the SQL type name NUM_ARRAY is created to store a VARRAY of NUMBER data. The NUM_ARRAY is in turn stored in a table VARRAY_TABLE. A query selects the contents of the VARRAY_TABLE. The result set is cast to OracleResultSet; getARRAY is applied to it to retrieve the array data into my_array, which is an oracle.sql.ARRAY object. Because my_array is of type oracle.sql.ARRAY, you can apply the methods getSQLTypeName and getBaseType to it to return the name of the SQL type of each element in the array and its integer code. The program then prints the contents of the array. Because the contents of NUM_ARRAY are of the SQL data type NUMBER, the elements of my_array are of the type, BigDecimal. Before you can use the elements, they must first be cast to BigDecimal. In the for loop, the individual values of the array are cast to BigDecimal and printed to standard output.
stmt.execute ("CREATE TYPE num_varray AS VARRAY(10) OF NUMBER(12, 2)"); stmt.execute ("CREATE TABLE varray_table (col1 num_varray)"); stmt.execute ("INSERT INTO varray_table VALUES (num_varray(100, 200))"); ResultSet rs = stmt.executeQuery("SELECT * FROM varray_table"); ARRAY my_array = ((OracleResultSet)rs).getARRAY(1); // return the SQL type names, integer codes, // and lengths of the columns System.out.println ("Array is of type " + array.getSQLTypeName());

Working with Oracle Collections 16-9

Creating and Using Arrays

System.out.println ("Array element is of type code " + array.getBaseType()); System.out.println ("Array is of length " + array.length()); // get Array elements BigDecimal[] values = (BigDecimal[]) my_array.getArray(); for (int i=0; i<values.length; i++) { BigDecimal out_value = (BigDecimal) values[i]; System.out.println(">> index " + i + " = " + out_value.intValue()); }

Note that if you use getResultSet to obtain the array, then you must would first get the result set object, and then use the next method to iterate through it. Notice the use of the parameter indexes in the getInt method to retrieve the element index and the element value.
ResultSet rset = my_array.getResultSet(); while (rset.next()) { // The first column contains the element index and the // second column contains the element value System.out.println(">> index " + rset.getInt(1)+" = " + rset.getInt(2)); }

Accessing Multilevel Collection Elements

The oracle.sql.ARRAY class provides three methods, which are overloaded, to access collection elements. The JDBC drivers extend these methods to support multilevel collections. These methods are:

getArray method getOracleArray method getResultSet method

The getArray method returns a Java array that holds the collection elements. The array element type is determined by the collection element type and the JDBC default conversion matrix. For example, the getArray method returns a java.math.BigDecimal array for collection of SQL NUMBER. The getOracleArray method returns a Datum array that holds the collection elements in Datum format. For multilevel collections, the getArray and getOracleArray methods both return a Java array of oracle.sql.ARRAY elements. The getResultSet method returns a ResultSet object that wraps the multilevel collection elements. For multilevel collections, the JDBC applications use the getObject, getARRAY, or getArray method of the ResultSet class to access the collection elements as instances of oracle.sql.ARRAY. The following code shows how to use the getOracleArray, getArray, and getResultSet methods:
Connection conn = ...; // make a JDBC connection Statement stmt = conn.createStatement (); ResultSet rset = stmt.executeQuery ("select col2 from tab2 where idx=1"); while (rset.next()) { ARRAY varray3 = (ARRAY) rset.getObject (1);

16-10 Oracle Database JDBC Developer's Guide and Reference

Creating and Using Arrays

Object varrayElems = varray3.getArray (1); // access array elements of "varray3" Datum[] varray3Elems = (Datum[]) varrayElems; for (int i=0; i<varray3Elems.length; i++) { ARRAY varray2 = (ARRAY) varray3Elems[i]; Datum[] varray2Elems = varray2.getOracleArray(); // access array elements of "varray2" for (int j=0; j<varray2Elems.length; j++) { ARRAY varray1 = (ARRAY) varray2Elems[j]; ResultSet varray1Elems = varray1.getResultSet(); // access array elements of "varray1" while (varray1Elems.next()) System.out.println ("idx="+varray1Elems.getInt(1)+" value="+varray1Elems.getInt(2)); } } } rset.close (); stmt.close (); conn.close ();

Passing Arrays to Statement Objects

This section discusses how to pass arrays to prepared statement objects or callable statement objects. Passing an Array to a Prepared Statement Pass an array to a prepared statement as follows.
Note:
1.

you can use arrays as either IN or OUT bind variables.

Define the array that you want to pass to the prepared statement as an oracle.sql.ARRAY object.
ARRAY array = oracle.jdbc.OracleConnection.createARRAY(sql_type_name, elements);

sql_type_name is a Java string specifying the user-defined SQL type name of the array and elements is a java.lang.Object containing a Java array of the elements.
2. 3.

Create a java.sql.PreparedStatement object containing the SQL statement to be run. Cast your prepared statement to OraclePreparedStatement, and use setARRAY to pass the array to the prepared statement.
(OraclePreparedStatement)stmt.setARRAY(parameterIndex, array);

parameterIndex is the parameter index and array is the oracle.sql.ARRAY object you constructed previously.
4.

Run the prepared statement.

Working with Oracle Collections 16-11

Using a Type Map to Map Array Elements

Passing an Array to a Callable Statement To retrieve a collection as an OUT parameter in PL/SQL blocks, perform the following to register the bind type for your OUT parameter.
1.

Cast your callable statement to OracleCallableStatement, as follows:

OracleCallableStatement ocs = (OracleCallableStatement)conn.prepareCall("{? = call func()}");

2.

Register the OUT parameter with the following form of the registerOutParameter method:
ocs.registerOutParameter (int param_index, int sql_type, string sql_type_name);

param_index is the parameter index, sql_type is the SQL type code, and sql_type_name is the name of the array type. In this case, the sql_type is OracleTypes.ARRAY.
3.

Run the call, as follows:

ocs.execute();

4.

Get the value, as follows:

oracle.sql.ARRAY array = ocs.getARRAY(1);

Using a Type Map to Map Array Elements

If your array contains Oracle objects, then you can use a type map to associate the objects in the array with the corresponding Java class. If you do not specify a type map, or if the type map does not contain an entry for a particular Oracle object, then each element is returned as an oracle.sql.STRUCT object. If you want the type map to determine the mapping between the Oracle objects in the array and their associated Java classes, then you must add an appropriate entry to the map. The following example illustrates how you can use a type map to map the elements of an array to a custom Java object class. In this case, the array is a nested table. The example begins by defining an EMPLOYEE object that has a name attribute and employee number attribute. EMPLOYEE_LIST is a nested table type of EMPLOYEE objects. Then an EMPLOYEE_TABLE is created to store the names of departments within a corporation and the employees associated with each department. In the EMPLOYEE_TABLE, the employees are stored in the form of EMPLOYEE_LIST tables.
stmt.execute("CREATE TYPE EMPLOYEE AS OBJECT (EmpName VARCHAR2(50),EmpNo INTEGER))"); stmt.execute("CREATE TYPE EMPLOYEE_LIST AS TABLE OF EMPLOYEE"); stmt.execute("CREATE TABLE EMPLOYEE_TABLE (DeptName VARCHAR2(20), Employees EMPLOYEE_LIST) NESTED TABLE Employees STORE AS ntable1"); stmt.execute("INSERT INTO EMPLOYEE_TABLE VALUES ("SALES", EMPLOYEE_LIST (EMPLOYEE('Susan Smith', 123), EMPLOYEE('Scott Tiger', 124)))");

If you want to retrieve all the employees belonging to the SALES department into an array of instances of the custom object class EmployeeObj, then you must add an entry to the type map to specify mapping between the EMPLOYEE SQL type and the EmployeeObj custom object class.
16-12 Oracle Database JDBC Developer's Guide and Reference

Custom Collection Classes with JPublisher

To do this, first create your statement and result set objects, then select the EMPLOYEE_LIST associated with the SALES department into the result set. Cast the result set to OracleResultSet so you can use the getARRAY method to retrieve the EMPLOYEE_LIST into an ARRAY object (employeeArray in the following example). The EmployeeObj custom object class in this example implements the SQLData interface.
Statement s = conn.createStatement(); OracleResultSet rs = (OracleResultSet)s.executeQuery ("SELECT Employees FROM employee_table WHERE DeptName = 'SALES'"); // get the array object ARRAY employeeArray = ((OracleResultSet)rs).getARRAY(1);

Now that you have the EMPLOYEE_LIST object, get the existing type map and add an entry that maps the EMPLOYEE SQL type to the EmployeeObj Java type.
// add type map entry to map SQL type // "EMPLOYEE" to Java type "EmployeeObj" Map map = conn.getTypeMap(); map.put("EMPLOYEE", Class.forName("EmployeeObj"));

Next, retrieve the SQL EMPLOYEE objects from the EMPLOYEE_LIST. To do this, call the getArray method of the employeeArray array object. This method returns an array of objects. The getArray method returns the EMPLOYEE objects into the employees object array.
// Retrieve array elements Object[] employees = (Object[]) employeeArray.getArray();

Finally, create a loop to assign each of the EMPLOYEE SQL objects to the EmployeeObj Java object emp.
// Each array element is mapped to EmployeeObj object. for (int i=0; i<employees.length; i++) { EmployeeObj emp = (EmployeeObj) employees[i]; ... }

Custom Collection Classes with JPublisher

This chapter primarily describes the functionality of the oracle.sql.ARRAY class, but it is also possible to access Oracle collections through custom Java classes or, more specifically, custom collection classes. You can create custom collection classes yourself, but the most convenient way is to use the Oracle JPublisher utility. Custom collection classes generated by JPublisher offer all the functionality described earlier in this chapter, as well as the following advantages:

They are strongly typed. This can help you find coding errors during compilation that might not otherwise be discovered until run time. They can be changeable, or mutable. Custom collection classes produced by JPublisher, unlike the ARRAY class, allow you to get and set individual elements using the getElement and setElement methods.

A custom collection class must satisfy three requirements:

Working with Oracle Collections 16-13

Custom Collection Classes with JPublisher

It must implement the oracle.sql.ORAData interface. Note that the standard JDBC SQLData interface, which is an alternative for custom object classes, is not intended for custom collection classes. It, or a companion class, must implement the oracle.sql.ORADataFactory interface, for creating instances of the custom collection class. It must have a means of storing the collection data. Typically it will directly or indirectly include an oracle.sql.ARRAY attribute for this purpose.

A JPublisher-generated custom collection class implements ORAData and ORADataFactory and indirectly includes an oracle.sql.ARRAY attribute. The custom collection class will have an oracle.jpub.runtime.MutableArray attribute. The MutableArray class has an oracle.sql.ARRAY attribute.
Note:

When you use JPublisher to create a custom collection class, you must use the ORAData implementation. This will be true if the JPublisher -usertypes mapping option is set to oracle, which is the default. You cannot use a SQLData implementation for a custom collection class. Setting the -usertypes mapping option to jdbc is invalid.

As an example of custom collection classes being strongly typed, if you define an Oracle collection MYVARRAY, then JPublisher can generate a MyVarray custom collection class. Using MyVarray instances, instead of generic oracle.sql.ARRAY instances, makes it easier to catch errors during compilation instead of at run time. For example, if you accidentally assign some other kind of array into a MyVarray variable. If you do not use custom collection classes, then you would use standard java.sql.Array instances, or oracle.sql.ARRAY instances, to map to your collections.
See Also:

"Using JPublisher to Create Custom Object Classes" on page 13-30 Oracle Database JPublisher User's Guide

16-14 Oracle Database JDBC Developer's Guide and Reference

17
Result Set
Standard Java Database Connectivity (JDBC) 2.0 features in Java Development Kit (JDK) 1.2.x include enhancements to result set functionality, such as processing forward or backward, positioning relatively or absolutely, seeing changes to the database made internally or externally, and updating result set data and then copying the changes to the database. This chapter discusses these features, including the following topics:

Overview Creating Scrollable or Updatable Result Sets Positioning and Processing in Scrollable Result Sets Updating Result Sets Fetch Size Refetching Rows Seeing Database Changes Made Internally and Externally

Overview
This section provides an overview of JDBC 2.0 result set functionality and categories, and some discussion of implementation requirements for Oracle JDBC drivers. This section covers the following topics:

Result Set Functionality and Result Set Categories Supported in JDBC 2.0 Oracle JDBC Implementation Overview for Result Set Enhancements

Result Set Functionality and Result Set Categories Supported in JDBC 2.0
Result set functionality in JDBC 2.0 includes enhancements for scrollability and positioning, sensitivity to changes by others, and updatability.

Scrollability, positioning, and sensitivity are determined by the result set type. Updatability is determined by the concurrency type.

Specify the desired result set type and concurrency type when you create the statement object that will produce the result set. Together, the various result set types and concurrency types provide for six different categories of result set.

Result Set 17-1

Overview

Scrollability, Positioning, and Sensitivity Scrollability refers to the ability to move backward as well as forward through a result set. Associated with scrollability is the ability to move to any particular position in the result set, through either relative positioning or absolute positioning. Relative positioning enables you to move a specified number of rows forward or backward from the current row. Absolute positioning enables you to move to a specified row number, counting from either the beginning or the end of the result set. Under JDBC 2.0, scrollable/positionable result sets are also available. When creating a scrollable/positionable result set, you must also specify sensitivity. This refers to the ability of a result set to detect and reveal changes made to the underlying database from outside the result set. A sensitive result set can see changes made to the database while the result set is open, providing a dynamic view of the underlying data. Changes made to the underlying columns values of rows in the result set are visible. An insensitive result set is not sensitive to changes made to the database while the result set is open, providing a static view of the underlying data. You would need to retrieve a new result set to see changes made to the database. Result Set Types for Scrollability and Sensitivity When you create a result set under JDBC 2.0 functionality, you must choose a particular result set type to specify whether the result set is scrollable/positional and sensitive to underlying database changes. If the JDBC 1.0 functionality is all you desire, JDBC 2.0 continues to support this through the forward-only result set type. A forward-only result set cannot be sensitive. If you want a scrollable result set, then you must also specify sensitivity. Specify the scroll-sensitive type for the result set to be scrollable and sensitive to underlying changes. Specify the scroll-insensitive type for the result set to be scrollable but not sensitive to underlying changes. To summarize, the following result set types are available with JDBC 2.0:

Forward-only This is a JDBC 1.0 functionality. This type of result set is not scrollable, not positionable, and not sensitive.

Scroll-sensitive This type of result set is scrollable and positionable. It is also sensitive to underlying database changes.

Scroll-insensitive This type of result set is scrollable and positionable, but not sensitive to underlying database changes.
Note:

The sensitivity of a scroll-sensitive result set is affected by fetch size.

Updatability Updatability refers to the ability to update data in a result set and then copy the changes to the database. This includes inserting new rows into the result set or deleting existing rows.

17-2 Oracle Database JDBC Developer's Guide and Reference

Overview

Updatability might also require database write locks to mediate access to the underlying database. Because you cannot have multiple write locks concurrently, updatability in a result set is associated with concurrency in database access. Result sets can optionally be updatable under JDBC 2.0
Note:

Updatability is independent of scrollability and sensitivity. Although, it is typical for an updatable result set to also be scrollable so that you can position it to particular rows that you want to update or delete.

Concurrency Types for Updatability The concurrency type of a result set determines whether it is updatable. Under JDBC 2.0, the following concurrency types are available:

Updatable In this case, updates, inserts, and deletes can be performed on the result set and copied to the database.

Read-only The result set cannot be modified in any way.

Summary of Result Set Categories Because scrollability and sensitivity are independent of updatability, the three result set types and two concurrency types combine for a total of six result set categories, as follows:

forward-only/read-only forward-only/updatable scroll-sensitive/read-only scroll-sensitive/updatable scroll-insensitive/read-only scroll-insensitive/updatable

Note:

A forward-only updatable result set has no positioning functionality. You can only update rows as you iterate through them with the next method.

Oracle JDBC Implementation Overview for Result Set Enhancements

This section discusses key aspects of the Oracle JDBC implementation of result set enhancements for scrollability, through use of a client-side cache, and for updatability, through use of ROWIDs. It is permissible for customers to implement their own client-side caching mechanism, and Oracle provides an interface to use in doing so. Oracle JDBC Implementation for Result Set Scrollability Because the underlying server does not support scrollable cursors, Oracle JDBC must implement scrollability in a separate layer.

Result Set 17-3

Overview

It is important to be aware that this is accomplished by using a client-side memory cache to store rows of a scrollable result set.
Important:

Because all rows of any scrollable result set are stored in the client-side cache, a situation where the result set contains many rows, many columns, or very large columns might cause the client-side Java Virtual Machine (JVM) to fail. Do not specify scrollability for a large result set.

Oracle JDBC Implementation for Result Set Updatability To support updatability, Oracle JDBC uses ROWID to uniquely identify database rows that appear in a result set. For every query into an updatable result set, Oracle JDBC driver automatically retrieves the ROWID along with the columns you select.
Note:

Client-side caching is not required by updatability in and of itself. In particular, a forward-only updatable result set will not require a client-side cache.

Implementing a Custom Client-Side Cache for Scrollability There is some flexibility in how to implement client-side caching in support of JDBC 2.0 scrollable result sets. Although Oracle JDBC provides a complete implementation, it also supplies an interface, OracleResultSetCache, that you can implement as desired:
public interface OracleResultSetCache { /** * Save the data in the i-th row and j-th column. */ public void put (int i, int j, Object value) throws IOException; /** * Return the data stored in the i-th row and j-th column. */ public Object get (int i, int j) throws IOException; /** * Remove the i-th row. */ public void remove (int i) throws IOException; /** * Remove the data stored in i-th row and j-th column */ public void remove (int i, int j) throws IOException; /** * Remove all data from the cache. */ public void clear () throws IOException; /** * Close the cache. */ public void close () throws IOException;

17-4 Oracle Database JDBC Developer's Guide and Reference

Creating Scrollable or Updatable Result Sets

If you implement this interface with your own class, then your application code must instantiate your class and then use the setResultSetCache method of an OracleStatement, OraclePreparedStatement, or OracleCallableStatement object to set the caching mechanism to use your implementation. Following is the method signature:
void setResultSetCache(OracleResultSetCache cache) throws SQLException

Call this method prior to running a query. The result set produced by the query will then use your specified caching mechanism.

Creating Scrollable or Updatable Result Sets

In using JDBC 2.0 result set enhancements, you may specify the result set type and the concurrency type when you create a generic statement or prepare a prepared statement or callable statement that will run a query. This section discusses the creation of result sets to use JDBC 2.0 enhancements. It covers the following topics:

Specifying Result Set Scrollability and Updatability Result Set Limitations and Downgrade Rules

Specifying Result Set Scrollability and Updatability

Under JDBC 2.0, the Connection class has the following methods that take a result set type and a concurrency type as input:

Statement createStatement(int resultSetType, int resultSetConcurrency) PreparedStatement prepareStatement(String sql, int resultSetType, int resultSetConcurrency) CallableStatement prepareCall(String sql, int resultSetType, int resultSetConcurrency)

The statement objects created will have the intelligence to produce the appropriate kind of result sets. You can specify one of the following static constant values for result set type:

ResultSet.TYPE_FORWARD_ONLY ResultSet.TYPE_SCROLL_INSENSITIVE ResultSet.TYPE_SCROLL_SENSITIVE

Tip: "Oracle Implementation of Scroll-Sensitive Result Sets" on page 17-19

And you can specify one of the following static constant values for concurrency type:

ResultSet.CONCUR_READ_ONLY ResultSet.CONCUR_UPDATABLE

Result Set 17-5

Creating Scrollable or Updatable Result Sets

After creating a Statement, PreparedStatement, or CallableStatement object, you can verify its result set type and concurrency type by calling the following methods on the statement object:

int getResultSetType() throws SQLException int getResultSetConcurrency() throws SQLException

Prepared Statement Object With Result Set

Example 171

Following is an example of a prepared statement object that specifies a scroll-sensitive and updatable result set for queries run through that statement:
... PreparedStatement pstmt = conn.prepareStatement ("SELECT empno, sal FROM emp WHERE empno = ?", ResultSet.TYPE_SCROLL_SENSITIVE, ResultSet.CONCUR_UPDATABLE); pstmt.setString(1, "28959"); ResultSet rs = pstmt.executeQuery(); ...

Result Set Limitations and Downgrade Rules

Some types of result sets are not feasible for certain kinds of queries. If you specify an unfeasible result set type or concurrency type for the query you run, then the JDBC driver follows a set of rules to determine the best feasible types to use instead. The actual result set type and concurrency type are determined when the statement is run, with the driver issuing a SQLWarning on the statement object if the desired result set type or concurrency type is not feasible. The SQLWarning object will contain the reason why the requested type was not feasible. Check for warnings to verify whether you received the type of result set that you requested. Result Set Limitations The following limitations are placed on queries for enhanced result sets. Failure to follow these guidelines will result in the JDBC driver choosing an alternative result set type or concurrency type. To produce an updatable result set:

A query can select from only a single table and cannot contain any join operations. In addition, for inserts to be feasible, the query must select all non-nullable columns and all columns that do not have a default value.

A query cannot use SELECT * . However, there is a workaround for this.

A query must select table columns only. It cannot select derived columns or aggregates, such as the SUM or MAX of a set of columns.

To produce a scroll-sensitive result set:

A query cannot use SELECT * . However, there is a workaround for this.

A query can select from only a single table.

17-6 Oracle Database JDBC Developer's Guide and Reference

Creating Scrollable or Updatable Result Sets

Scrollable and updatable result sets cannot have any column as Stream. When the server has to fetch a Stream column, it reduces the fetch size to one and blocks all columns following the Stream column until the Stream column is read. As a result, columns cannot be fetched in bulk and scrolled through. Workaround As a workaround for the SELECT * limitation, you can use table aliases, as shown in the following example:
SELECT t.* FROM TABLE t ...

Note:

There is a simple way to determine if your query will probably produce a scroll-sensitive or updatable result set: If you can legally add a ROWID column to the query list, then the query is probably suitable for either a scroll-sensitive or an updatable result set.

Result Set Downgrade Rules If the specified result set type or concurrency type is not feasible, then Oracle JDBC driver uses the following rules in choosing alternate types:

If the specified result set type is TYPE_SCROLL_SENSITIVE, but the JDBC driver cannot fulfill that request, then the driver attempts a downgrade to TYPE_SCROLL_INSENSITIVE. If the specified or downgraded result set type is TYPE_SCROLL_INSENSITIVE, but the JDBC driver cannot fulfill that request, then the driver attempts a downgrade to TYPE_FORWARD_ONLY. If the specified concurrency type is CONCUR_UPDATABLE, but the JDBC driver cannot fulfill that request, then the JDBC driver attempts a downgrade to CONCUR_READ_ONLY.
Note:

Any manipulations of the result set type and concurrency type by the JDBC driver are independent of each other.

Verifying Result Set Type and Concurrency Type After a query has been run, you can verify the result set type and concurrency type that the JDBC driver actually used, by calling methods on the result set object.

int getType() throws SQLException This method returns an int value for the result set type used for the query. ResultSet.TYPE_FORWARD_ONLY, ResultSet.TYPE_SCROLL_SENSITIVE, or ResultSet.TYPE_SCROLL_INSENSITIVE are the possible values.

int getConcurrency() throws SQLException This method returns an int value for the concurrency type used for the query. ResultSet.CONCUR_READ_ONLY or ResultSet.CONCUR_UPDATABLE are the possible values.

Result Set 17-7

Positioning and Processing in Scrollable Result Sets

Positioning and Processing in Scrollable Result Sets

Scrollable result sets enable you to iterate through them, either forward or backward, and to position the result set to any desired row. This section discusses positioning within a scrollable result set and how to process a scrollable result set backward, instead of forward. It covers the following sections:

Positioning in a Scrollable Result Set

In a scrollable result set, you can use several result set methods to move to a desired position and to check the current position.
Note:

You cannot position a forward-only result set. Any attempt to position it or to determine the current position will result in a SQLException exception.

Methods for Moving to a New Position The following result set methods are available for moving to a new position in a scrollable result set:

void beforeFirst() throws SQLException Positions to before the first row of the result set, or has no effect if there are no rows in the result set. This is where you would typically start iterating through a result set to process it going forward and is the default initial position for any kind of result set. You are outside the result set bounds after a beforeFirst() call. There is no valid current row, and you cannot position relatively from this point.

void afterLast() throws SQLException Positions to after the last row of the result set, or has no effect if there are no rows in the result set. This is where you would typically start iterating through a result set to process it going backward. You are outside the result set bounds after an afterLast() call. There is no valid current row, and you cannot position relatively from this point.

boolean first() throws SQLException Positions to the first row of the result set, or returns false if there are no rows in the result set.

boolean last() throws SQLException Positions to the last row of the result set, or returns false if there are no rows in the result set.

boolean absolute(int row) throws SQLException Positions to an absolute row from either the beginning or end of the result set. If you enter a positive number, then it positions from the beginning. If you enter a negative number, then it positions from the end. This method returns false if there are no rows in the result set. Attempting to move forward beyond the last row, such as an absolute(11) call if there are 10 rows, will position to after the last row, having the same effect as an afterLast() call.

17-8 Oracle Database JDBC Developer's Guide and Reference

Positioning and Processing in Scrollable Result Sets

Attempting to move backward beyond the first row, such as an absolute(-11) call if there are 10 rows, will position to before the first row, having the same effect as a beforeFirst() call.
Note: Calling absolute(1) is equivalent to calling first(); calling absolute(-1) is equivalent to calling last().

boolean relative(int row) throws SQLException Moves to a position relative to the current row, forward if you enter a positive number or backward if you enter a negative number, or returns false if there are no rows in the result set. The result set must be at a valid current row for use of the relative method. Attempting to move forward beyond the last row will position to after the last row, having the same effect as an afterLast() call. Attempting to move backward beyond the first row will position to before the first row, having the same effect as a beforeFirst() call. A relative(0) call is valid but has no effect.
Note:

You cannot position relatively from before the first row, which is the default initial position, or after the last row. Attempting relative positioning from either of these positions would result in a SQLException exception.

Methods for Checking the Current Position The following result set methods are available for checking the current position in a scrollable result set:

boolean isBeforeFirst() throws SQLException Returns true if the position is before the first row.

boolean isAfterLast() throws SQLException Returns true if the position is after the last row.

boolean isFirst() throws SQLException Returns true if the position is at the first row.

boolean isLast() throws SQLException Returns true if the position is at the last row.

int getRow() throws SQLException Returns the row number of the current row, or returns 0 if there is no valid current row. The boolean methods, isFirst(), isLast(), isAfterFirst(), and isAfterLast(), all return false. Also, they do not throw an exception if there are no rows in the result set.
Note:

Result Set 17-9

Positioning and Processing in Scrollable Result Sets

Processing a Scrollable Result Set

In a scrollable result set you can iterate backward instead of forward as you process the result set. The following methods are available:

boolean next() throws SQLException boolean previous() throws SQLException

The previous() method works similarly to the next() method, in that it returns true as long as the new current row is valid, and false as soon as it runs out of rows, that is, has passed the first row. Backward versus Forward Processing You can process the entire result set going forward, using the next() method. The default initial position in the result set is before the first row, appropriately, but you can call the beforeFirst() method if you have moved elsewhere since the result set was created. To process the entire result set going backward, call afterLast(), then use the previous() method. For example:
... /* NOTE: The specified concurrency type, CONCUR_UPDATABLE, is not relevant to this example. */ Statement stmt = conn.createStatement (ResultSet.TYPE_SCROLL_SENSITIVE, ResultSet.CONCUR_UPDATABLE); ResultSet rs = stmt.executeQuery("SELECT empno, sal FROM emp"); rs.afterLast(); while (rs.previous()) { System.out.println(rs.getString("empno") + " " + rs.getFloat("sal")); } ...

Unlike relative positioning, you can use next() from before the first row and previous() from after the last row. You do not have to be at a valid current row to use these methods.
Note:

In a non-scrollable result set, you can process only with the next() method. Attempting to use the previous() method will cause a SQLException exception.

Presetting the Fetch Direction The JDBC 2.0 standard allows the ability to pre-specify the direction, known as the fetch direction, for use in processing a result set. This allows the JDBC driver to optimize its processing. The following result set methods are specified:

void setFetchDirection(int direction) throws SQLException int getFetchDirection() throws SQLException

Oracle JDBC drivers support only the forward preset value, which you can specify by entering the ResultSet.FETCH_FORWARD static constant value.

17-10 Oracle Database JDBC Developer's Guide and Reference

Updating Result Sets

The values ResultSet.FETCH_REVERSE and ResultSet.FETCH_UNKNOWN are not supported. Attempting to specify them causes a SQL warning, and the settings are ignored.

Updating Result Sets

A concurrency type of CONCUR_UPDATABLE enables you to update rows in the result set, delete rows from the result set, or insert rows into the result set. After you perform an UPDATE or INSERT operation in a result set, you propagate the changes to the database in a separate step that you can skip if you want to cancel the changes. However, a DELETE operation in a result set is immediately run, but not necessarily committed, in the database as well.
Note:

When using an updatable result set, it is typical to also make it scrollable. This enables you to position to any row that you want to change. With a forward-only updatable result set, you can change rows only as you iterate through them with the next() method.

This section covers the following topics:

Performing a DELETE Operation in a Result Set Performing an UPDATE Operation in a Result Set Performing an INSERT Operation in a Result Set Avoiding Update Conflicts

Performing a DELETE Operation in a Result Set

The result set deleteRow method will delete the current row. Following is the method signature:
void deleteRow() throws SQLException

Note:

Unlike UPDATE and INSERT operations in a result set, which require a separate step to propagate the changes to the database, a DELETE operation in a result set is immediately run in the corresponding row in the database as well. Once you call deleteRow method, the changes are made permanent with the next transaction COMMIT operation. Remember also that by default, the auto-commit flag is set to true. Therefore, unless you override this default, any deleteRow method operation is run and committed immediately.

Presuming the result set is also scrollable, you can position to a row using any of the available positioning methods, except beforeFirst method and afterLast method, which do not go to a valid current row, and then delete that row, as in the following example:
...

Result Set 17-11

Updating Result Sets

rs.absolute(5); rs.deleteRow(); ...

Important:

The deleted row remains in the result set object even after it has been deleted from the database. In a scrollable result set, by contrast, a DELETE operation is evident in the local result set object. The row would no longer be in the result set after the DELETE. The row preceding the deleted row becomes the current row, and row numbers of subsequent rows are changed accordingly.

Performing an UPDATE Operation in a Result Set

Performing a result set UPDATE operation requires two separate steps to first update the data in the result set and then copy the changes to the database. Presuming the result set is also scrollable, you can position to a row using any of the available positioning methods, except beforeFirst() and afterLast(), which do not go to a valid current row, and then update that row as desired. Here are the steps for updating a row in the result set and database:
1.

Call the appropriate updateXXX methods to update the data in the columns you want to change. With JDBC 2.0, a result set object has an updateXXX method for each data type, as with the setXXX methods previously available for updating the database directly. Each of these methods takes an int for the column number or a string for the column name and then an item of the appropriate data type to set the new value. Following are a couple of examples for a result set rs:
rs.updateString(1, "mystring"); rs.updateFloat(2, 10000.0f);

2.

Call the updateRow method to copy the changes to the database or the cancelRowUpdates method to cancel the changes. Once you call the updateRow method, the changes are run and are made permanent with the next transaction COMMIT operation. Be aware that by default, the auto-commit flag is set to true so that any operation run is committed immediately. If you choose to cancel the changes before copying them to the database, then call the cancelRowUpdates method instead. This will also revert to the original values for that row in the local result set object. Note that once you call the updateRow method, the changes are written to the transaction and cannot be canceled unless you roll back the transaction.
Note:

Auto-commit must be disabled to allow a ROLLBACK operation.

Positioning to a different row before calling updateRow also cancels the changes and reverts to the original values in the result set.

17-12 Oracle Database JDBC Developer's Guide and Reference

Updating Result Sets

Before calling updateRow, you can call the usual getXXX methods to verify that the values have been updated correctly. These methods take an int column index or string column name as input. For example:
float myfloat = rs.getFloat(2); ... // process myfloat to see if it's appropriate ...

Note:

Result set UPDATE operations are visible in the local result set object for all result set types, forward-only, scroll-sensitive, and scroll-insensitive.

Example Following is an example of a result set UPDATE operation that is also copied to the database. The tenth row is updated. The column number is used to specify column 1, and the column name, sal, is used to specify column 2.
Statement stmt = conn.createStatement (ResultSet.TYPE_SCROLL_SENSITIVE, ResultSet.CONCUR_UPDATABLE); ResultSet rs = stmt.executeQuery("SELECT empno, sal FROM emp"); if (rs.absolute(10)) // (returns false if row does not exist) { rs.updateString(1, "28959"); rs.updateFloat("sal", 100000.0f); rs.updateRow(); } // Changes are made permanent with the next COMMIT operation.

Performing an INSERT Operation in a Result Set

Result set INSERT operations use what is called the result set insert-row, which is a staging area that holds the data for the inserted row until it is copied to the database. You must explicitly move to this row to write the data that will be inserted. As with UPDATE operations, result set INSERT operations require separate steps to first write the data to the insert-row and then copy it to the database. Following are the steps in running a result set INSERT operation.
1.

Move to the insert-row by calling the result set moveToInsertRow method.

Note:

The result set will remember the current position prior to the moveToInsertRow call. Afterward, you can go back to it with a moveToCurrentRow call.

2.

As with UPDATE operations, use the appropriate updateXXX methods to write data to the columns. For example:
rs.updateString(1, "mystring"); rs.updateFloat(2, 10000.0f);

You can specify a string for column name, instead of an integer for column number.

Result Set 17-13

Updating Result Sets

Important:

Each column value in the insert-row is undefined until you call the updateXXX method for that column. You must call this method and specify a non-null value for all non-nullable columns, or else attempting to copy the row into the database will result in a SQLException exception.

However, it is permissible to not call updateXXX for a nullable column. This will result in a value of null.
3.

Copy the changes to the database by calling the result set insertRow method. Once you call insertRow, the insert is processed and will be made permanent with the next transaction COMMIT operation. Positioning to a different row before calling insertRow cancels the insert and clears the insert-row. Before calling insertRow you can call the usual getXXX methods to verify that the values have been set correctly in the insert-row. These methods take an int column index or string column name as input. For example:
float myfloat = rs.getFloat(2); ...process myfloat to see if it's appropriate...

Note:

No result set type can see a row inserted by a result set INSERT operation.

Example The following example performs a result set INSERT operation, moving to the insert-row, writing the data, copying the data into the database, and then returning to what was the current row prior to going to the insert-row. The column number is used to specify column 1, and the column name, sal, is used to specify column 2.
... Statement stmt = conn.createStatement (ResultSet.TYPE_SCROLL_SENSITIVE, ResultSet.CONCUR_UPDATABLE); ResultSet rs = stmt.executeQuery("SELECT empno, sal FROM emp"); rs.moveToInsertRow(); rs.updateString(1, "28959"); rs.updateFloat("sal", 100000.0f); rs.insertRow(); // Changes will be made permanent with the next COMMIT operation. rs.moveToCurrentRow(); // Go back to where we came from... ...

Avoiding Update Conflicts

It is important to be aware of the following facts regarding updatable result sets with the JDBC drivers:

The drivers do not enforce write locks for an updatable result set. The drivers do not check for conflicts with a result set DELETE or UPDATE operation.

17-14 Oracle Database JDBC Developer's Guide and Reference

Fetch Size

A conflict will occur if you try to perform a DELETE or UPDATE operation on a row updated by another committed transaction. Oracle JDBC drivers use the ROWID to uniquely identify a row in a database table. As long as the ROWID is valid when a driver tries to send an UPDATE or DELETE operation to the database, the operation will be run. The driver will not report any changes made by another committed transaction. Any conflicts are silently ignored and your changes will overwrite the previous changes. To avoid such conflicts, use the Oracle FOR UPDATE feature when running the query that produces the result set. This will avoid conflicts, but will also prevent simultaneous access to the data. Only a single write lock can be held concurrently on a data item.

Fetch Size
By default, when Oracle JDBC runs a query, it retrieves a result set of 10 rows at a time from the database cursor. This is the default Oracle row fetch size value. You can change the number of rows retrieved with each trip to the database cursor by changing the row fetch size value. JDBC 2.0 also enables you to specify the number of rows fetched with each database round-trip for a query, and this number is referred to as the fetch size. In Oracle JDBC, the row-prefetch value is used as the default fetch size in a statement object. Setting the fetch size overrides the row-prefetch setting and affects subsequent queries run through that statement object. Fetch size is also used in a result set. When the statement object run a query, the fetch size of the statement object is passed to the result set object produced by the query. However, you can also set the fetch size in the result set object to override the statement fetch size that was passed to it.
Note:

Changes made to the fetch size of a statement object after a result set is produced will have no affect on that result set.

The result set fetch size, either set explicitly, or by default equal to the statement fetch size that was passed to it, determines the number of rows that are retrieved in any subsequent trips to the database for that result set. This includes any trips that are still required to complete the original query, as well as any refetching of data into the result set. Data can be refetched, either explicitly or implicitly, to update a scroll-sensitive or scroll-insensitive/updatable result set.

Setting the Fetch Size

The following methods are available in all Statement, PreparedStatement, CallableStatement, and ResultSet objects for setting and getting the fetch size:

void setFetchSize(int rows) throws SQLException int getFetchSize() throws SQLException

To set the fetch size for a query, call setFetchSize on the statement object prior to running the query. If you set the fetch size to N, then N rows are fetched with each trip to the database. After you have run the query, you can call setFetchSize on the result set object to override the statement object fetch size that was passed to it. This will affect any

Result Set 17-15

Refetching Rows

subsequent trips to the database to get more rows for the original query, as well as affecting any later refetching of rows.

Refetching Rows
The result set refreshRow method is supported for some types of result sets for refetching data. This consists of going back to the database to re-obtain the database rows that correspond to n rows in the result set, starting with the current row, where n is the fetch size. This lets you see the latest updates to the database that were made outside of your result set, subject to the isolation level of the enclosing transaction. Because refetching re-obtains only rows that correspond to rows already in your result set, it does nothing about rows that have been inserted or deleted in the database since the original query. It ignores rows that have been inserted, and rows will remain in your result set even after the corresponding rows have been deleted from the database. When there is an attempt to refetch a row that has been deleted in the database, the corresponding row in the result set will maintain its original values.
Note:

If you declare a TYPE_SCROLL_SENSITIVE resultset based on a query with certain criteria and then externally update the row so that the column values no longer match the query criteria, the driver behaves as if the row has been deleted from the database and the row is not retrieved by the query issued. So, you do not see the updates to the particular row when you call the refreshRow method.

Following is the signature of the refreshRow method:

void refreshRow() throws SQLException

You must be at a valid current row when you call this method, not outside the row bounds and not at the insert-row. The refreshRow method is supported for the following result set categories:

scroll-sensitive/read-only scroll-sensitive/updatable scroll-insensitive/updatable

Note:

Scroll-sensitive result set functionality is implemented through implicit calls to refreshRow.

Seeing Database Changes Made Internally and Externally

This section discusses the ability of a result set to see the following:

Its own changes, referred to as internal changes Changes made from elsewhere, either from your own transaction outside the result set, or from other committed transactions, referred to as external changes
Note:

External changes are referred to as other's changes in the Sun Microsystems JDBC 2.0 specification.

17-16 Oracle Database JDBC Developer's Guide and Reference

Seeing Database Changes Made Internally and Externally

This section covers the following topics:

Seeing Internal Changes Seeing External Changes Visibility versus Detection of External Changes Summary of Visibility of Internal and External Changes Oracle Implementation of Scroll-Sensitive Result Sets

Seeing Internal Changes

The ability of an updatable result set to see its own changes depends on both the result set type and the kind of change. This is summarized as follows:

Internal DELETE operations are visible for scrollable result sets, but are not visible for forward-only result sets. After you delete a row in a scrollable result set, the preceding row becomes the new current row, and subsequent row numbers are updated accordingly.

Internal UPDATE operations are always visible, regardless of the result set type. Internal INSERT operations are never visible, regardless of the result set type.

An internal change being visible essentially means that a subsequent getXXX call will see the data changed by a preceding updateXXX call on the same data item. JDBC 2.0 DatabaseMetaData objects include the following methods to verify this:

boolean ownDeletesAreVisible(int) throws SQLException boolean ownUpdatesAreVisible(int) throws SQLException boolean ownInsertsAreVisible(int) throws SQLException

Each takes a result set type, ResultSet.TYPE_FORWARD_ONLY, ResultSet.TYPE_SCROLL_SENSITIVE, or ResultSet.TYPE_SCROLL_INSENSITIVE, as input.

Note:

When you make an internal change that causes a trigger to run, the trigger changes are effectively external changes. However, if the trigger affects data in the row you are updating, then you will see those changes for any scrollable/updatable result set, because an implicit row refetch occurs after the update.

Seeing External Changes

Only a scroll-sensitive result set can see external changes to the underlying database, and it can only see the changes from external UPDATE operations. Changes from external DELETE or INSERT operations are never visible.
Note:

Any discussion of seeing changes from outside the enclosing transaction presumes the transaction itself has an isolation level setting that allows the changes to be visible.

JDBC 2.0 DatabaseMetaData objects include the following methods to verify this:

boolean othersDeletesAreVisible(int) throws SQLException

Result Set 17-17

Seeing Database Changes Made Internally and Externally

boolean othersUpdatesAreVisible(int) throws SQLException boolean othersInsertsAreVisible(int) throws SQLException

Each takes a result set type, ResultSet.TYPE_FORWARD_ONLY, ResultSet.TYPE_SCROLL_SENSITIVE, or ResultSet.TYPE_SCROLL_INSENSITIVE, as input.

Note:

Explicit use of the refreshRow method is distinct from this discussion of visibility. For example, even though external updates are invisible to a scroll-insensitive result set, you can explicitly refetch rows in a scroll-insensitive/updatable result set and retrieve external changes that have been made. Visibility refers only to the fact that the scroll-insensitive/updatable result set would not see such changes automatically and implicitly.

Visibility versus Detection of External Changes

Regarding changes made to the underlying database by external sources, there are two similar but distinct concepts with respect to visibility of the changes from your local result set:

Visibility of changes Detection of changes

A change being "visible" means that when you look at a row in the result set, you can see new data values from changes made by external sources to the corresponding row in the database. A change being "detected", however, means that the result set is aware that this is a new value since the result set was first populated. Even when an Oracle result set sees new data, as with an external UPDATE in a scroll-sensitive result set, it has no awareness that this data has changed since the result set was populated. Such changes are not detected. JDBC 2.0 DatabaseMetaData objects include the following methods to verify this:

boolean deletesAreDetected(int) throws SQLException boolean updatesAreDetected(int) throws SQLException boolean insertsAreDetected(int) throws SQLException

Each takes a result set type, ResultSet.TYPE_FORWARD_ONLY, ResultSet.TYPE_SCROLL_SENSITIVE, or ResultSet.TYPE_SCROLL_INSENSITIVE, as input. It follows, then, that result set methods specified by JDBC 2.0 to detect changes, rowDeleted, rowUpdated, and rowInserted, will always return false. There is no use in calling them.

Summary of Visibility of Internal and External Changes

Table 171 summarizes the discussion in the preceding sections regarding whether a result set object in the Oracle JDBC implementation can see changes made internally through the result set itself, and changes made externally to the underlying database from elsewhere in your transaction or from other committed transactions.

17-18 Oracle Database JDBC Developer's Guide and Reference

Seeing Database Changes Made Internally and Externally

Table 171

Visibility of Internal and External Changes for Oracle JDBC Can See Can See Can See Internal Internal Internal DELETE? UPDATE? INSERT? no yes yes yes yes yes no no no Can See Can See External External DELETE? UPDATE? no no no no yes no Can See External INSERT? no no no

Result Set Type forward-only scroll-sensitive scroll-insensitive

Note:

Remember that explicit use of the refreshRow method, is distinct from the concept of visibility of external changes. Remember that even when external changes are visible, as with UPDATE operations underlying a scroll-sensitive result set, they are not detected. The result set rowDeleted, rowUpdated, and rowInserted methods always return false.

Oracle Implementation of Scroll-Sensitive Result Sets

The Oracle implementation of scroll-sensitive result sets involves the concept of a window, with a window size that is based on the fetch size. The window size affects how often rows are updated in the result set. Once you establish a current row by moving to a specified row, the window consists of the n rows in the result set starting with that row, where n is the fetch size being used by the result set. Note that there is no current row, and therefore no window, when a result set is first created. The default position is before the first row, which is not a valid current row. As you move from row to row, the window remains unchanged as long as the current row stays within that window. However, once you move to a new current row outside the window, you redefine the window to be the N rows starting with the new current row. Whenever the window is redefined, the N rows in the database corresponding to the rows in the new window are automatically refetched through an implicit call to the refreshRow method, thereby updating the data throughout the new window. So external updates are not instantaneously visible in a scroll-sensitive result set. They are only visible after the automatic refetches just described.

Note:

Because this kind of refetching is not a highly efficient or optimized methodology, there are significant performance concerns. Consider carefully before using scroll-sensitive result sets as currently implemented. There is also a significant trade-off between sensitivity and performance. The most sensitive result set is one with a fetch size of 1, which would result in the new current row being refetched every time you move between rows. However, this would have a significant impact on the performance of your application.

Result Set 17-19

Seeing Database Changes Made Internally and Externally

17-20 Oracle Database JDBC Developer's Guide and Reference

18
JDBC RowSets
This chapter contains the following sections:

Overview CachedRowSet JdbcRowSet WebRowSet FilteredRowSet JoinRowSet

Overview
A RowSet is an object that encapsulates a set of rows from either java Database Connectivity (JDBC) result sets or tabular data sources. RowSets support component-based development models like JavaBeans, with a standard set of properties and an event notification mechanism. RowSets were introduced in JDBC 2.0 through the optional packages. However, the implementation of RowSets was standardized in the JDBC RowSet Implementations Specification (JSR-114), which is available as non-optional package since Java Platform, Standard Edition (Java SE) 5.0. Java SE 6.0 RowSets contain more APIs supporting features like RowId, National Language Charactersets, and so on. The Java SE Javadocs provide information about the standard interfaces and base classes for JDBC RowSet implementations.
See Also:

JSR-114 specification at: http://jcp.org/en/jsr/detail?id=114 Java SE 5.0 Javadoc at: http://java.sun.com/j2se/1.5.0/docs/api/ Java SE 6.0 Javadoc at: http://java.sun.com/javase/6/docs/api/

Note:

In case of any conflict, the JSR-114 specification takes precedence over the JDK 5.0 Javadoc.

The JSR-114 specification includes implementation details for five types of RowSet:

JDBC RowSets 18-1

Overview

CachedRowSet JdbcRowSet WebRowSet FilteredRowSet JoinRowSet

Oracle JDBC supports all five types of RowSets through the interfaces and classes present in the oracle.jdbc.rowset package. Oracle Database 11g Release 1 (11.1) adds RowSets support in the server-side drivers. Therefore, starting from Oracle Database 11g Release 1 (11.1), RowSets support is uniform across all Oracle JDBC driver types. The standard Oracle JDBC Java Archive (JAR) files, for example, ojdbc5.jar and ojdbc6.jar contain the oracle.jdbc.rowset package.
Note:

The other JAR files with different file suffix names, for example, ojdbc5_g.jar, ojdbc5dms.jar, and so on also contain the oracle.jdbc.rowset package. In Oracle Database 10g release 2 (10.2), the implementation classes were packaged in the ojdbc14.jar file. Prior to Oracle Database 10g release 2 (10.2), the implementation classes were packaged in the ocrs12.jar file. Prior to Oracle Database 11g Release 1 (11.1), RowSets support was not available in the server-side drivers.

To use the Oracle RowSet implementations, you need to import either the entire oracle.jdbc.rowset package or specific classes and interfaces from the package for the required RowSet type. For client-side usage, you also need to include the standard Oracle JAR files like ojdbc5.jar or ojdbc6.jar in the CLASSPATH environment variable.
See Also:

"Check the Environment Variables" on page 2-3 for information about setting the CLASSPATH environment variable.

This section covers the following topics:

RowSet Properties Events and Event Listeners Command Parameters and Command Execution Traversing RowSets

RowSet Properties
The javax.sql.RowSet interface provides a set of JavaBeans properties that can be altered to access the data in the data source through a single interface. Example of properties are connection string, user name, password, type of connection, and the query string. For a complete list of properties and property descriptions, refer to the Java2 Platform, Standard Edition (J2SE) Javadoc for javax.sql.RowSet at http://java.sun.com/j2se/1.5.0/docs/api/javax/sql/RowSet.html

18-2 Oracle Database JDBC Developer's Guide and Reference

Overview

The interface provides standard accessor methods for setting and retrieving the property values. The following code illustrates setting some of the RowSet properties:
... rowset.setUrl("jdbc:oracle:oci:@"); rowset.setUsername("SCOTT"); rowset.setPassword("TIGER"); rowset.setCommand("SELECT empno, ename, sal FROM emp"); ...

In this example, the URL, user name, password, and SQL query are set as the RowSet properties to retrieve the employee number, employee name, and salary of all the employees into the RowSet object.

Events and Event Listeners

RowSets support JavaBeans events. The following types of events are supported by the RowSet interface:

cursorMoved This event is generated whenever there is a cursor movement. For example, when the next or previous method is called.

rowChanged This event is generated when a row is inserted, updated, or deleted from the RowSet.

rowSetChanged This event is generated when the whole RowSet is created or changed. For example, when the execute method is called.

An application component can implement a RowSet listener to listen to these RowSet events and perform desired operations when the event occurs. Application components, which are interested in these events, must implement the standard javax.sql.RowSetListener interface and register such listener objects with a RowSet object. A listener can be registered using the RowSet.addRowSetListener method and unregistered using the RowSet.removeRowSetListener method. Multiple listeners can be registered with the same RowSet object. The following code illustrates the registration of a RowSet listener:
... MyRowSetListener rowsetListener = new MyRowSetListener (); // adding a rowset listener rowset.addRowSetListener (rowsetListener); ...

The following code illustrates a listener implementation:

public class MyRowSetListener implements RowSetListener { public void cursorMoved(RowSetEvent event) { // action on cursor movement } public void rowChanged(RowSetEvent event) { // action on change of row }

JDBC RowSets 18-3

Overview

public void rowSetChanged(RowSetEvent event) { // action on changing of rowset } }// end of class MyRowSetListener

Applications that need to handle only selected events can implement only the required event handling methods by using the oracle.jdbc.rowset.OracleRowSetListenerAdapter class, which is an abstract class with empty implementation for all the event handling methods. In the following code, only the rowSetChanged event is handled, while the remaining events are not handled by the application:
... rowset.addRowSetListener(new oracle.jdbc.rowset.OracleRowSetListenerAdapter () { public void rowSetChanged(RowSetEvent event) { // your action for rowSetChanged } } ); ...

Command Parameters and Command Execution

The command property of a RowSet object typically represents a SQL query string, which when processed would populate the RowSet object with actual data. Like in regular JDBC processing, this query string can take input or bind parameters. The javax.sql.RowSet interface also provides methods for setting input parameters to this SQL query. After the required input parameters are set, the SQL query can be processed to populate the RowSet object with data from the underlying data source. The following code illustrates this simple sequence:
... rowset.setCommand("SELECT ename, sal FROM emp WHERE empno = ?"); // setting the employee number input parameter for employee named "KING" rowset.setInt(1, 7839); rowset.execute(); ...

In the preceding example, the employee number 7839 is set as the input or bind parameter for the SQL query specified in the command property of the RowSet object. When the SQL query is processed, the RowSet object is filled with the employee name and salary information of the employee whose employee number is 7839.

Traversing RowSets
The javax.sql.RowSet interface extends the java.sql.ResultSet interface. The RowSet interface, therefore, provides cursor movement and positioning methods, which are inherited from the ResulSet interface, for traversing through data in a RowSet object. Some of the inherited methods are absolute, beforeFirst, afterLast, next, and previous. The RowSet interface can be used just like a ResultSet interface for retrieving and updating data. The RowSet interface provides an optional way to implement a

18-4 Oracle Database JDBC Developer's Guide and Reference

Overview

scrollable and updatable result set. All the fields and methods provided by the ResultSet interface are implemented in RowSet. The Oracle implementation of ResultSet provides the scrollable and updatable properties of the java.sql.ResultSet interface.
Note:

The following code illustrates how to scroll through a RowSet:

/** * Scrolling forward, and printing the empno in * the order in which it was fetched. */ ... rowset.setCommand("SELECT empno, ename, sal FROM emp"); rowset.execute(); ... // going to the first row of the rowset rowset.beforeFirst (); while (rowset.next ()) System.out.println ("empno: " +rowset.getInt (1));

In the preceding code, the cursor position is initialized to the position before the first row of the RowSet by the beforeFirst method. The rows are retrieved in forward direction using the next method. The following code illustrates how to scroll through a RowSet in the reverse direction:
/** * Scrolling backward, and printing the empno in * the reverse order as it was fetched. */ //going to the last row of the rowset rowset.afterLast (); while (rowset.previous ()) System.out.println ("empno: " +rowset.getInt (1));

In the preceding code, the cursor position is initialized to the position after the last row of the RowSet. The rows are retrieved in reverse direction using the previous method of RowSet. Inserting, updating, and deleting rows are supported by the RowSet feature as they are in the ResultSet feature. In order to make the RowSet updatable, you must call the setReadOnly(false) and acceptChanges methods. The following code illustrates the insertion of a row at the fifth position of a RowSet:
... /** * Make rowset updatable */ rowset.setReadOnly (false); /** * Inserting a row in the 5th position of the rowset. */ // moving the cursor to the 5th position in the rowset if (rowset.absolute(5)) { rowset.moveToInsertRow (); rowset.updateInt (1, 193);

JDBC RowSets 18-5

CachedRowSet

rowset.updateString (2, "Ashok"); rowset.updateInt (3, 7200); // inserting a row in the rowset rowset.insertRow (); // Synchronizing the data in RowSet with that in the database. rowset.acceptChanges (); } ...

In the preceding code, a call to the absolute method with a parameter 5 takes the cursor to the fifth position of the RowSet and a call to the moveToInsertRow method creates a place for the insertion of a new row into the RowSet. The updateXXX methods are used to update the newly created row. When all the columns of the row are updated, the insertRow is called to update the RowSet. The changes are committed through acceptChanges method.

CachedRowSet
A CachedRowSet is a RowSet in which the rows are cached and the RowSet is disconnected, that is, it does not maintain an active connection to the database. The oracle.jdbc.rowset.OracleCachedRowSet class is the Oracle implementation of CachedRowSet. It can interoperate with the reference implementation of Sun Microsystems. The OracleCachedRowSet class in the ojdbc5.jar and ojdbc6.jar files implements the standard JSR-114 interface javax.sql.rowset.CachedRowSet. In the following code, an OracleCachedRowSet object is created and the connection URL, user name, password, and the SQL query for the RowSet object is set as properties. The RowSet object is populated using the execute method. After the execute method has been processed, the RowSet object can be used as a java.sql.ResultSet object to retrieve, scroll, insert, delete, or update data.
... RowSet rowset = new OracleCachedRowSet(); rowset.setUrl("jdbc:oracle:oci:@"); rowset.setUsername("SCOTT"); rowset.setPassword("TIGER"); rowset.setCommand("SELECT empno, ename, sal FROM emp"); rowset.execute(); while (rowset.next ()) { System.out.println("empno: " +rowset.getInt (1)); System.out.println("ename: " +rowset.getString (2)); System.out.println("sal: " +rowset.getInt (3)); } ...

To populate a CachedRowSet object with a query, complete the following steps:

1. 2.

Instantiate OracleCachedRowSet. Set the Url, which is the connection URL, Username, Password, and Command, which is the query string, properties for the RowSet object. You can also set the connection type, but it is optional. Call the execute method to populate the CachedRowSet object. Calling execute runs the query set as a property on this RowSet.

3.

18-6 Oracle Database JDBC Developer's Guide and Reference

CachedRowSet

OracleCachedRowSet rowset = new OracleCachedRowSet (); rowset.setUrl ("jdbc:oracle:oci:@"); rowset.setUsername ("SCOTT"); rowset.setPassword ("TIGER"); rowset.setCommand ("SELECT empno, ename, sal FROM emp"); rowset.execute ();

A CachedRowSet object can be populated with an existing ResultSet object, using the populate method. To do so, complete the following steps:
1. 2.

Instantiate OracleCachedRowSet. Pass the already available ResultSet object to the populate method to populate the RowSet object.
// Executing a query to get the ResultSet object. ResultSet rset = pstmt.executeQuery (); OracleCachedRowSet rowset = new OracleCachedRowSet (); // the obtained ResultSet object is passed to the populate method // to populate the data in the rowset object. rowset.populate (rset);

In the preceding example, a ResultSet object is obtained by running a query and the retrieved ResultSet object is passed to the populate method of the CachedRowSet object to populate the contents of the result set into the CachedRowSet.
Note:

Connection properties, like transaction isolation or the concurrency mode of the result set, and the bind properties cannot be set in the case where a pre-existent ResultSet object is used to populate the CachedRowSet object, because the connection or result set on which the property applies would have already been created.

The following code illustrates how an OracleCachedRowSet object is serialized to a file and then retrieved:
// writing the serialized OracleCachedRowSet object { FileOutputStream fileOutputStream = new FileOutputStream("emp_tab.dmp"); ObjectOutputStream ostream = new ObjectOutputStream(fileOutputStream); ostream.writeObject(rowset); ostream.close(); fileOutputStream.close(); } // reading the serialized OracleCachedRowSet object { FileInputStream fileInputStream = new FileInputStream("emp_tab.dmp"); ObjectInputStream istream = new ObjectInputStream(fileInputStream); RowSet rowset1 = (RowSet) istream.readObject(); istream.close(); fileInputStream.close(); }

In the preceding code, a FileOutputStream object is opened for an emp_tab.dmp file, and the populated OracleCachedRowSet object is written to the file using

JDBC RowSets 18-7

CachedRowSet

ObjectOutputStream. The serialized OracleCachedRowSet object is retrieved using the FileInputStream and ObjectInputStream objects. OracleCachedRowSet takes care of the serialization of non-serializable form of data like InputStream, OutputStream, binary large objects (BLOBs), and character large objects (CLOBs). OracleCachedRowSets also implements metadata of its own, which could be obtained without any extra server round-trip. The following code illustrates how you can obtain metadata for the RowSet:
... ResultSetMetaData metaData = rowset.getMetaData(); int maxCol = metaData.getColumnCount(); for (int i = 1; i <= maxCol; ++i) System.out.println("Column (" + i +") " + metaData.getColumnName(i)); ...

Because the OracleCachedRowSet class is serializable, it can be passed across a network or between Java Virtual Machines (JVMs), as done in Remote Method Invocation (RMI). Once the OracleCachedRowSet class is populated, it can move around any JVM, or any environment that does not have JDBC drivers. Committing the data in the RowSet requires the presence of JDBC drivers. The complete process of retrieving the data and populating it in the OracleCachedRowSet class is performed on the server and the populated RowSet is passed on to the client using suitable architectures like RMI or Enterprise Java Beans (EJB). The client would be able to perform all the operations like retrieving, scrolling, inserting, updating, and deleting on the RowSet without any connection to the database. Whenever data is committed to the database, the acceptChanges method is called, which synchronizes the data in the RowSet to that in the database. This method makes use of JDBC drivers, which require the JVM environment to contain JDBC implementation. This architecture would be suitable for systems involving a Thin client like a Personal Digital Assistant (PDA). After populating the CachedRowSet object, it can be used as a ResultSet object or any other object, which can be passed over the network using RMI or any other suitable architecture. Some of the other key-features of CachedRowSet are the following:

Cloning a RowSet Creating a copy of a RowSet Creating a shared copy of a RowSet

CachedRowSet Constraints All the constraints that apply to an updatable result set are applicable here, except serialization, because OracleCachedRowSet is serializable. The SQL query has the following constraints:

References only a single table in the database Contains no join operations Selects the primary key of the table it references

In addition, a SQL query should also satisfy the following conditions, if new rows are to be inserted:

Selects all non-nullable columns in the underlying table Selects all columns that do not have a default value

18-8 Oracle Database JDBC Developer's Guide and Reference

JdbcRowSet

Note: The CachedRowSet cannot hold a large quantity of data, because all the data is cached in memory. Oracle, therefore, recommends against using OracleCachedRowSet with queries that could potentially return a large volume of data.

Connection properties like, transaction isolation and concurrency mode of the result set, cannot be set after populating the RowSet, because the properties cannot be applied to the connection after retrieving the data from the same.

JdbcRowSet
A JdbcRowSet is a RowSet that wraps around a ResultSet object. It is a connected RowSet that provides JDBC interfaces in the form of a JavaBean interface. The Oracle implementation of JdbcRowSet is oracle.jdbc.rowset.OracleJDBCRowSet. The OracleJDBCRowSet class in ojdbc5.jar and ojdbc6.jar implements the standard JSR-114 interface javax.sql.rowset.JdbcRowSet. Table 181 shows how the JdbcRowSet interface differs from CachedRowSet interface.
Table 181 The JDBC and Cached Row Sets Compared Connected Movable to Across Database JVMs Yes No No Yes Synchronization Presence of data to of JDBC database Drivers No Yes Yes No

RowSet Type JDBC Cached

Serializable Yes Yes

JdbcRowSet is a connected RowSet, which has a live connection to the database and all the calls on the JdbcRowSet are percolated to the mapping call in the JDBC connection, statement, or result set. A CachedRowSet does not have any connection to the database open. JdbcRowSet requires the presence of JDBC drivers unlike a CachedRowSet, which does not require JDBC drivers during manipulation. However, both JdbcRowSet and CachedRowSet require JDBC drivers during population of the RowSet and while committing the changes of the RowSet. The following code illustrates how a JdbcRowSet is used:
... RowSet rowset = new OracleJDBCRowSet(); rowset.setUrl("java:oracle:oci:@"); rowset.setUsername("SCOTT"); rowset.setPassword("TIGER"); rowset.setCommand("SELECT empno, ename, sal FROM emp"); rowset.execute(); while (rowset.next()) { System.out.println("empno: " + rowset.getInt(1)); System.out.println("ename: " + rowset.getString(2)); System.out.println("sal: " + rowset.getInt(3)); } ...

JDBC RowSets 18-9

WebRowSet

In the preceding example, the connection URL, user name, password, and SQL query are set as properties of the RowSet object, the SQL query is processed using the execute method, and the rows are retrieved and printed by traversing through the data populated in the RowSet object.

WebRowSet
A WebRowSet is an extension to CachedRowSet. It represents a set of fetched rows or tabular data that can be passed between tiers and components in a way such that no active connections with the data source need to be maintained. The WebRowSet interface provides support for the production and consumption of result sets and their synchronization with the data source, both in Extensible Markup Language (XML) format and in disconnected fashion. This allows result sets to be shipped across tiers and over Internet protocols. The Oracle implementation of WebRowSet is oracle.jdbc.rowset.OracleWebRowSet. This class, which is in the ojdbc5.jar and ojdbc6.jar files, implements the standard JSR-114 interface javax.sql.rowset.WebRowSet. This class also extends the oracle.jdbc.rowset.OracleCachedRowSet class. Besides the methods available in OracleCachedRowSet, the OracleWebRowSet class provides the following methods:
public OracleWebRowSet() throws SQLException

This is the constructor for creating an OracleWebRowSet object, which is initialized with the default values for an OracleCachedRowSet object, a default OracleWebRowSetXmlReader, and a default OracleWebRowSetXmlWriter.
public void writeXml(java.io.Writer writer) throws SQLException public void writeXml(java.io.OutputStream ostream) throws SQLException

These methods write the OracleWebRowSet object to the supplied Writer or OutputStream object in the XML format that conforms to the JSR-114 XML schema. In addition to the RowSet data, the properties and metadata of the RowSet are written.
public void writeXml(ResultSet rset, java.io.Writer writer) throws SQLException public void writeXml(ResultSet rset, java.io.OutputStream ostream) throws SQLException

These methods create an OracleWebRowSet object, populate it with the data in the given ResultSet object, and write it to the supplied Writer or OutputStream object in the XML format that conforms to the JSR-114 XML schema.
public void readXml(java.io.Reader reader) throws SQLException public void readXml(java.io.InputStream istream) throws SQLException

These methods read the OracleWebRowSet object in the XML format according to its JSR-114 XML schema, using the supplied Reader or InsputStream object. The Oracle WebRowSet implementation supports Java API for XML Processing (JAXP) 1.2. Both Simple API for XML (SAX) 2.0 and Document Object Model (DOM) JAXP-conforming XML parsers are supported. It follows the current JSR-114 W3C XML schema for WebRowSet from Sun Microsystems, which is at: http://java.sun.com/xml/ns/jdbc/webrowset.xsd

18-10 Oracle Database JDBC Developer's Guide and Reference

WebRowSet

See Also:

JSR-114 specification at: http://jcp.org/en/jsr/detail?id=114 Java SE 5.0 Javadoc at: http://java.sun.com/j2se/1.5.0/docs/api/ Java SE 6.0 Javadoc at: http://java.sun.com/javase/6/docs/api/

Applications that use the readXml(...) methods should set one of the following two standard JAXP system properties before calling the methods:

javax.xml.parsers.SAXParserFactory This property is for a SAX parser.

javax.xml.parsers.DocumentBuilderFactory This property is for a DOM parser.

The following code illustrates the use of OracleWebRowSet for both writing and reading in XML format:
import java.sql.*; import java.io.*; import oracle.jdbc.rowset.*; ... String url = "jdbc:oracle:oci8:@"; Connection conn = DriverManager.getConnection(url,"scott","tiger"); Statement stmt = conn.createStatement(); ResultSet rset = stmt.executeQuery("select * from emp"); // Create an OracleWebRowSet object and populate it with the ResultSet object OracleWebRowSet wset = new OracleWebRowSet(); wset.populate(rset); try { // Create a java.io.Writer object FileWriter out = new FileWriter("xml.out"); // Now generate the XML and write it out wset.writeXml(out); } catch (IOException exc) { System.out.println("Couldnt construct a FileWriter"); } System.out.println("XML output file generated."); // Create a new OracleWebRowSet for reading from XML input OracleWebRowSet wset2 = new OracleWebRowSet(); // Use Oracle JAXP SAX parser System.setProperty("javax.xml.parsers.SAXParserFactory","oracle.xml.jaxp.JXSAXPars erFactory"); try

JDBC RowSets

18-11

FilteredRowSet

{ // Use the preceding output file as input FileReader fr = new FileReader("xml.out"); // Now read XML stream from the FileReader wset2.readXml(fr); } catch (IOException exc) { System.out.println("Couldnt construct a FileReader"); } ...

Note:

The preceding code uses the Oracle SAX XML parser, which supports schema validation.

FilteredRowSet
A FilteredRowSet is an extension to WebRowSet that provides programmatic support for filtering its content. This enables you to avoid the overhead of supplying a query and the processing involved. The Oracle implementation of FilteredRowSet is oracle.jdbc.rowset.OracleFilteredRowSet. The OracleFilteredRowSet class in the ojdbc5.jar and ojdbc6.jar files implements the standard JSR-114 interface javax.sql.rowset.FilteredRowSet. The OracleFilteredRowSet class defines the following new methods:
public Predicate getFilter();

This method returns a Predicate object that defines the filtering criteria active on the OracleFilteredRowSet object.
public void setFilter(Predicate p) throws SQLException;

This method takes a Predicate object as a parameter. The Predicate object defines the filtering criteria to be applied on the OracleFilteredRowSet object. The methods throws a SQLException exception. The predicate set on an OracleFilteredRowSet object defines a filtering criteria that is applied to all the rows in the object to obtain the set of visible rows. The predicate also defines the criteria for inserting, deleting, and modifying rows. The set filtering criteria acts as a gating mechanism for all views and updates to the OracleFilteredRowSet object. Any attempt to update the OracleFilteredRowSet object, which violates the filtering criteria, throws a SQLException exception. The filtering criteria set on an OracleFilteredRowSet object can be modified by applying a new Predicate object. The new criteria is immediately applied on the object, and all further views and updates must adhere to this new criteria. A new filtering criteria can be applied only if there are no reference to the OracleFilteredRowSet object. Rows that fall outside of the filtering criteria set on the object cannot be modified until the filtering criteria is removed or a new filtering criteria is applied. Also, only the rows that fall within the bounds of the filtering criteria will be synchronized with the data source, if an attempt is made to persist the object.

18-12 Oracle Database JDBC Developer's Guide and Reference

JoinRowSet

The following code example illustrates the use of OracleFilteredRowSet. Assume a table, test_table, with two NUMBER columns, col1 and col2. The code retrieves those rows from the table that have value of col1 between 50 and 100 and value of col2 between 100 and 200. The predicate defining the filtering criteria is as follows:
public class PredicateImpl implements Predicate { private int low[]; private int high[]; private int columnIndexes[]; public PredicateImpl(int[] lo, int[] hi, int[] indexes) { low = lo; high = hi; columnIndexes = indexes; } public boolean evaluate(RowSet rs) { boolean result = true; for (int i = 0; i < columnIndexes.length; i++) { int columnValue = rs.getInt(columnIndexes[i]); if (columnValue < low[i] || columnValue > high[i]) result = false; } return result; } // the other two evaluate(...) methods simply return true }

The predicate defined in the preceding code is used for filtering content in an OracleFilteredRowSet object, as follows:
... OracleFilteredRowSet ofrs = new OracleFilteredRowSet(); int low[] = {50, 100}; int high[] = {100, 200}; int indexes[] = {1, 2}; ofrs.setCommand("select col1, col2 from test_table"); // set other properties on ofrs like usr/pwd ... ... ofrs.execute(); ofrs.setPredicate(new PredicateImpl(low, high, indexes)); // this will only get rows with col1 in (50,100) and col2 in (100,200) while (ofrs.next()) {...} ...

JoinRowSet
A JoinRowSet is an extension to WebRowSet that consists of related data from different RowSets. There is no standard way to establish a SQL JOIN between

JDBC RowSets

18-13

JoinRowSet

disconnected RowSets without connecting to the data source. A JoinRowSet addresses this issue. The Oracle implementation of JoinRowSet is the oracle.jdbc.rowset.OracleJoinRowSet class. This class, which is in the ojdbc5.jar and ojdbc6.jar files, implements the standard JSR-114 interface javax.sql.rowset.JoinRowSet. Any number of RowSet objects, which implement the Joinable interface, can be added to a JoinRowSet object, provided they can be related in a SQL JOIN. All five types of RowSet support the Joinable interface. The Joinable interface provides methods for specifying the columns based on which the JOIN will be performed, that is, the match columns. A match column can be specified in the following ways:

Using the setMatchColumn method This method is defined in the Joinable interface. It is the only method that can be used to set the match column before a RowSet object is added to a JoinRowSet object. This method can also be used to reset the match column at any time.

Using the addRowSet method This is an overloaded method in JoinRowSet. Four of the five implementations of this method take a match column as a parameter. These four methods can be used to set or reset a match column at the time a RowSet object is being added to a JoinRowSet object.

In addition to the inherited methods, OracleJoinRowSet provides the following methods:

public public public public public void void void void void addRowSet(Joinable joinable) throws SQLException; addRowSet(RowSet rowSet, int i) throws SQLException; addRowSet(RowSet rowSet, String s) throws SQLException; addRowSet(RowSet arowSet[], int an[]) throws SQLException; addRowSet(RowSet arowSet[], String as[]) throws SQLException;

These methods are used to add a RowSet object to the OracleJoinRowSet object. You can pass one or more RowSet objects to be added to the OracleJoinRowSet object. You can also pass names or indexes of one or more columns, which need to be set as match column.
public Collection getRowSets() throws SQLException;

This method retrieves the RowSet objects added to the OracleJoinRowSet object. The method returns a java.util.Collection object that contains the RowSet objects.
public String[] getRowSetNames() throws SQLException;

This method returns a String array containing the names of the RowSet objects that are added to the OracleJoinRowSet object.
public public public public public boolean boolean boolean boolean boolean supportsCrossJoin(); supportsFullJoin(); supportsInnerJoin(); supportsLeftOuterJoin(); supportsRightOuterJoin();

These methods return a boolean value indicating whether the OracleJoinRowSet object supports the corresponding JOIN type.
public void setJoinType(int i) throws SQLException; 18-14 Oracle Database JDBC Developer's Guide and Reference

JoinRowSet

This method is used to set the JOIN type on the OracleJoinRowSet object. It takes an integer constant as defined in the javax.sql.rowset.JoinRowSet interface that specifies the JOIN type.
public int getJoinType() throws SQLException;

This method returns an integer value that indicates the JOIN type set on the OracleJoinRowSet object. This method throws a SQLException exception.
public CachedRowSet toCachedRowSet() throws SQLException;

This method creates a CachedRowSet object containing the data in the OracleJoinRowSet object.
public String getWhereClause() throws SQLException;

This method returns a String containing the SQL-like description of the WHERE clause used in the OracleJoinRowSet object. This methods throws a SQLException exception. The following code illustrates how OracleJoinRowSet is used to perform an inner join on two RowSets, whose data come from two different tables. The resulting RowSet contains data as if they were the result of an inner join on these two tables. Assume that there are two tables, an Order table with two NUMBER columns Order_id and Person_id, and a Person table with a NUMBER column Person_id and a VARCHAR2 column Name.
... // RowSet holding data from table Order OracleCachedRowSet ocrsOrder = new OracleCachedRowSet(); ... ocrsOrder.setCommand("select order_id, person_id from order"); ... // Join on person_id column ocrsOrder.setMatchColumn(2); ocrsOrder.execute(); // Creating the JoinRowSet OracleJoinRowSet ojrs = new OracleJoinRowSet(); ojrs.addRowSet(ocrsOrder); // RowSet holding data from table Person OracleCachedRowSet ocrsPerson = new OracleCachedRowSet(); ... ocrsPerson.setCommand("select person_id, name from person"); ... // do not set match column on this RowSet using setMatchColumn(). //use addRowSet() to set match column ocrsPerson.execute(); // Join on person_id column, in another way ojrs.addRowSet(ocrsPerson, 1); // now we can go the JoinRowSet as usual ojrs.beforeFirst(); while (ojrs.next()) System.out.println("order id = " + ojrs.getInt(1) + ", " + "person id = " + ojrs.getInt(2) + ", " + "person's name = " + ojrs.getString(3)); ...

JDBC RowSets

18-15

JoinRowSet

18-16 Oracle Database JDBC Developer's Guide and Reference

19
Globalization Support
The Oracle Java Database Connectivity (JDBC) drivers provide globalization support, formerly known as National Language Support (NLS). Globalization support enables you retrieve data or insert data into a database in any character set that Oracle supports. If the clients and the server use different character sets, then the driver provides the support to perform the conversions between the database character set and the client character set. This chapter contains the following sections:

Providing Globalization Support NCHAR, NVARCHAR2, NCLOB and the defaultNChar Property in JDK 1.5 New Methods for National Character Set Type Data in JDK 1.6
See Also:

"Oracle Character Data Types Support" on page 4-10 Oracle Database Globalization Support Guide Oracle Database Reference

Note:

Starting from Oracle Database 10g, the NLS_LANG variable is no longer part of the JDBC globalization mechanism. The JDBC driver does not check NLS environment. So, setting it has no effect. The JDBC server-side internal driver provides complete globalization support and does not require any globalization extension files. JDBC 4.0 includes methods for reading and writing national character set values. You should use these methods when using JSE 6 or later.

Providing Globalization Support

The basic Java Archive (JAR) files, ojdbc5.jar and ojdbc6.jar, contain all the necessary classes to provide complete globalization support for:

Oracle character sets for CHAR, VARCHAR, LONGVARCHAR, or CLOB data that is not being retrieved or inserted as a data member of an Oracle object or collection type.

Globalization Support 19-1

Providing Globalization Support

CHAR or VARCHAR data members of object and collection for the character sets US7ASCII, WE8DEC, WE8ISO8859P1, WE8MSWIN1252, and UTF8.

To use any other character sets in CHAR or VARCHAR data members of objects or collections, you must include orai18n.jar in the CLASSPATH environment variable of your application.
Note:

Previous releases depended on the nls_charset12.zip file. This file is now obsolete.

Compressing orai18n.jar The orai18n.jar file contains many important character set and globalization support files. You can reduce the size of orai18n.jar using the built-in customization tool, as follows:
java -jar orai18n.jar -custom-charsets-jar [jar/zip_filename] -charset characterset_name [characterset_name ...]

For example, if you want to create a custom character set file, custom_orai18n_ja.jar, that includes the JA16SJIS and JA16EUC character sets, then issue the following command:
$ java -jar orai18n.jar -custom-charsets-jar custom_orai18n_ja.jar -charset JA16SJIS JA16EUC

The output of the command is as follows:

Added Character set : JA16SJIS Added Character set : JA16EUC

If you do not specify a file name for your custom JAR/ZIP file, then a file with the name jdbc_orai18n_cs.jar is created in the current working directory. Also, for your custom JAR/ZIP file, you cannot specify a name that starts with orai18n. If any invalid or unsupported character set name is specified in the command, then no output JAR/ZIP file will be created. If the custom JAR/ZIP file exists, then the file will not be updated or removed. The custom character set JAR/ZIP does not accept any command. However, it prints the version information and the command that was used to generate the JAR/ZIP file. For example, you have jdbc_orai18n_cs.zip, the command that displays the information and the displayed information is as follows:
$ java -jar jdbc_orai18n_cs.jar Oracle Globalization Development Kit - 10.2.X.X.X Release This custom character set jar/zip file was created with the following command: java -jar orai18n.jar -custom-charsets-jar jdbc_orai18n_cs.jar -charset WE8ISO8859P15

The limitation to the number of character sets that can be specified depends on that of the shell or command prompt of the operating system. It is certified that all supported character sets can be specified with the command.

19-2 Oracle Database JDBC Developer's Guide and Reference

NCHAR, NVARCHAR2, NCLOB and the defaultNChar Property in JDK 1.5

Note:
1.

If you are using a custom character set, then you need to perform the following so that JDBC supports the custom character set:
After creating the .nlt and .nlb files as part of the process of creating a custom character set, create .glb files for the newly created character set and also for the lx0boot.nlt file using the following command: java -classpath $ORACLE_HOME/jlib/orai18n-tools.jar Ginstall <nlt file>

2.

Add the generated files and $ORACLE_HOME/jlib/orai18n-mappings.jar into the classpath environment variable while executing the JDBC code that connects to the database with the custom character set.

For more information about creating a custom character set, refer to Oracle Database Globalization Support Guide.

NCHAR, NVARCHAR2, NCLOB and the defaultNChar Property in JDK 1.5

By default, oracle.jdbc.OraclePreparedStatement treats all columns as CHAR. However, since Oracle Database 10g, if you set the system property oracle.jdbc.defaultNChar to true, then JDBC treats all character columns as being national-language. The default value for defaultNChar is false. To set defaultNChar, you specify the following at the command-line:
java -Doracle.jdbc.defaultNChar=true myApplication

If you prefer, then your application can specify defaultNChar as a connection property and access NCHAR, NVARCHAR2, or NCLOB data. For example:
PreparedStatement pstmt = conn.prepareStatement("insert into TEST values(?,?,?)"); pstmt.setInt(1, 1); // NUMBER column pstmt.setString(2, myUnicodeString1); // NVARCHAR2 column pstmt.setString(3, myUnicodeString2); // NCHAR column pstmt.execute();

However, if you set defaultNChar to true and then access CHAR columns, then the database will implicitly convert all CHAR data into NCHAR. This conversion has a substantial performance impact. To avoid this, use the setNString method to access NCHAR and NVARCHAR2 data and setNClob method to access NCLOB data. For example:
PreparedStatement pstmt = conn.prepareStatement("insert into TEST values(?,?,?)"); pstmt.setInt(1, 1); // NUMBER column pstmt.setNString(2, myUnicodeString1); // NVARCHAR2 column pstmt.setNString(3, myUnicodeString2); // NCHAR column pstmt.setString(4, myString); // CHAR column pstmt.execute();

Note: You can also use the setObject method to access national character set types, but if the setObject method is used , then the target data type must be specified as Types.NCHAR, Types.NCLOB, Types.NVARCHAR, or Types.LONGNVARCHAR.

Globalization Support 19-3

New Methods for National Character Set Type Data in JDK 1.6

Note:

In Oracle Database, SQL strings are converted to the database character set. Therefore you need to keep in mind the following: In Oracle Database 10g release 1 (10.1) and earlier releases, JDBC drivers do not support any NCHAR literal (n'...') containing Unicode characters that are not representable in the database character set. All Unicode characters that are not representable in the database character set get corrupted. If an Oracle Database 10g release 2 (10.2) JDBC driver is connected to an Oracle Database 10g release 2 (10.2) database server, then all NCHAR literals (n'...') are converted to Unicode literals (u'...') and all non-ASCII characters are converted to their corresponding Unicode escape sequence. This is done automatically to prevent data corruption. If an Oracle Database 10g release 2 (10.2) JDBC driver is connected to an Oracle Database 10g release 1 (10.1) or earlier database server, then NCHAR literals (n'...') are not converted and any character that is not representable in the database character set gets corrupted.

New Methods for National Character Set Type Data in JDK 1.6
JDBC 4.0 introduces new JDBC data types, such as NCHAR, NVARCHAR, LONGNVARCHAR, and NCLOB to access the national character set types. These types are similar to the CHAR, VARCHAR, LONGVARCHAR, and CLOB types, except that the values are encoded using the national character set. The JDBC specification uses Strings to represent NCHAR, NVARCHAR, and LONGNVARCHAR data, and NClob to represent NCLOB values. To retrieve a national character value, an application calls one of the following methods:

getString getClob getCharacterStream getObject

Note: The getClob method may be used to return an NClob object since NClob implements Clob.

To specify a value for a parameter marker of national character type, an application calls one of the following methods:

setNString setNCharacterStream setNClob setObject

Tip:

If the setObject method is used , then the target data type must be specified as Types.NCHAR, Types.NCLOB, Types.NVARCHAR, or Types.LONGNVARCHAR.

19-4 Oracle Database JDBC Developer's Guide and Reference

Part V
Performance and Scalability
This part consists of chapters that discuss the Oracle Java Database Connectivity (JDBC) features that enhance performance, such as statement caching, implicit connection caching, run-time connection load balancing, and Oracle Call Interface (OCI) connection pooling. It also includes a chapter that provides information about Oracle performance extensions, such as update batching and row prefetching. Part V contains the following chapters:

Chapter 20, "Statement and Resultset Caching" Chapter 21, "Implicit Connection Caching" Chapter 22, "Run-Time Connection Load Balancing" Chapter 23, "Performance Extensions" Chapter 24, "OCI Connection Pooling" Chapter 25, "Oracle Advanced Queuing"

20
Statement and Resultset Caching
This chapter describes the benefits and use of statement caching, an Oracle Java Database Connectivity (JDBC) extension. This chapter contains the following sections:

About Statement Caching Using Statement Caching Reusing Statements Objects Resultset Caching
Note: In Oracle9i Database 9.2.0 and later releases, Oracle JDBC provides a new statement cache interface and implementation, replacing the application programming interface (API) supported in Oracle9i Database release 1 (9.0.1). The previous API is now deprecated.

About Statement Caching

Statement caching improves performance by caching executable statements that are used repeatedly, such as in a loop or in a method that is called repeatedly. Starting from JDBC 3.0, JDBC standards define a statement-caching interface. Statement caching can:

Prevent the overhead of repeated cursor creation Prevent repeated statement parsing and creation Reuse data structures in the client

This section covers the following topics:

Basics of Statement Caching Implicit Statement Caching Explicit Statement Caching

Note:

Oracle strongly recommends you to use the implicit Statement cache. Oracle JDBC drivers are designed on the assumption that the implicit statement cache is enabled. So, not using the Statement cache will have a negative impact on performance.

Statement and Resultset Caching

20-1

About Statement Caching

Basics of Statement Caching

Applications use the statement cache to cache statements associated with a particular physical connection. The cache is associated with an OracleConnection object. OracleConnection includes methods to enable statement caching. When you enable statement caching, a statement object is cached when you call the close method. Because each physical connection has its own cache, multiple caches can exist if you enable statement caching for multiple physical connections. When you enable statement caching on a connection cache, the logical connections benefit from the statement caching that is enabled on the underlying physical connection. If you try to enable statement caching on a logical connection held by a connection cache, then this will throw an exception. There are two types of statement caching: implicit and explicit. Each type of statement cache can be enabled or disabled independent of the other. You can have either, neither, or both in effect. Both types of statement caching share a single cache per connection.

Implicit Statement Caching

When you enable implicit statement caching, JDBC automatically caches the prepared or callable statement when you call the close method of this statement object. The prepared and callable statements are cached and retrieved using standard connection object and statement object methods. Plain statements are not implicitly cached, because implicit statement caching uses a SQL string as a key and plain statements are created without a SQL string. Therefore, implicit statement caching applies only to the OraclePreparedStatement and OracleCallableStatement objects, which are created with a SQL string. You cannot use implicit statement caching with OracleStatement. When you create an OraclePreparedStatement or OracleCallableStatement, the JDBC driver automatically searches the cache for a matching statement. The match criteria are the following:

The SQL string in the statement must be identical to one in the cache. The statement type must be the same, that is, prepared or callable. The scrollable type of result sets produced by the statement must be the same, that is, forward-only or scrollable.

If a match is found during the cache search, then the cached statement is returned. If a match is not found, then a new statement is created and returned. In either case, the statement, along with its cursor and state are cached when you call the close method of the statement object. When a cached OraclePreparedStatement or OracleCallableStatement object is retrieved, the state and data information are automatically re-initialized and reset to default values, while metadata is saved. Statements are removed from the cache to conform to the maximum size using an Least Recently Used (LRU) algorithm.
Note:

The JDBC driver does not clear metadata. However, although metadata is saved for performance reasons, it has no semantic impact. A statement that comes from the implicit cache appears as if it were newly created.

You can prevent a particular statement from being implicitly cached.

20-2 Oracle Database JDBC Developer's Guide and Reference

Using Statement Caching

See Also: "Disabling Implicit Statement Caching for a Particular Statement" on page 20-6

Explicit Statement Caching

Explicit statement caching enables you to cache and retrieve selected prepared and callable statements. Explicit statement caching relies on a key, an arbitrary Java String that you provide.
Note:

Plain statements cannot be cached.

Because explicit statement caching retains statement data and state as well as metadata, it has a performance edge over implicit statement caching, which retains only metadata. However, you must be cautious when using this type of caching, because explicit statement caching saves all three types of information for reuse and you may not be aware of what data and state are retained from prior use of the statements. Implicit and explicit statement caching can be differentiated on the following points:

Retrieving statements In the case of implicit statement caching, you take no special action to retrieve statements from a cache. Instead, whenever you call prepareStatement or prepareCall, JDBC automatically checks the cache for a matching statement and returns it if found. However, in the case of explicit statement caching, you use specialized Oracle WithKey methods to cache and retrieve statement objects.

Providing key Implicit statement caching uses the SQL string of a prepared or callable statement as the key, requiring no action on your part. In contrast, explicit statement caching requires you to provide a Java String, which it uses as the key.

Returning statements During implicit statement caching, if the JDBC driver cannot find a statement in cache, then it will automatically create one. However, during explicit statement caching, if the JDBC driver cannot find a matching statement in cache, then it will return a null value.

Table 201 compares the different methods employed in implicit and explicit statement caching.
Table 201 Comparing Methods Used in Statement Caching Insert Into Cache close closeWithKey Retrieve From Cache prepareStatement prepareCall getStatementWithKey getCallWithKey

Allocate Implicit Explicit prepareStatement prepareCall createStatement prepareStatement prepareCall

Using Statement Caching

This section discusses the following topics:

Enabling and Disabling Statement Caching

Statement and Resultset Caching

20-3

Using Statement Caching

Closing a Cached Statement Using Implicit Statement Caching Using Explicit Statement Caching

Enabling and Disabling Statement Caching

When using the OracleConnection API, implicit and explicit statement caching can be enabled or disabled independent of one other. You can have either, neither, or both in effect. Enabling Implicit Statement Caching Enable implicit statement caching in one of the following two ways:

Call setImplicitCachingEnabled(true) on the connection Call OracleDataSource.getConnection with the ImplicitCachingEnabled property set to true. You set ImplicitCachingEnabled by calling OracleDataSource.setImplicitCachingEnabled(true)

In addition to calling one of these methods, you also need to call OracleConnection.setStatementCacheSize on the physical connection. The argument you supply is the maximum number of statements in the cache. An argument of 0 specifies no caching. To determine whether implicit caching is enabled, call getImplicitCachingEnabled, which returns true if implicit caching is enabled, false otherwise.
Note:

Enabling statement caching enables both implicit and explicit statement caching.

Disabling Implicit Statement Caching Disable implicit statement caching by calling setImplicitCachingEnabled(false) on the connection or by setting the ImplicitCachingEnabled property to false. Enabling Explicit Statement Caching To enable explicit statement caching you must first set the statement cache size. For setting the cache size, call OracleConnection.setStatementCacheSize method on the physical connection. The argument you supply is the maximum number of statements in the cache. An argument of 0 specifies no caching. To check the cache size, use the getStatementCacheSize method in the following way:
System.out.println("Stmt Cache size is " + ((OracleConnection)conn).getStatementCacheSize());

The following code specifies a cache size of ten statements:

((OracleConnection)conn).setStatementCacheSize(10);

Enable explicit statement caching by calling setExplicitCachingEnabled(true) on the connection.

20-4 Oracle Database JDBC Developer's Guide and Reference

Using Statement Caching

To determine whether explicit caching is enabled, call getExplicitCachingEnabled, which returns true if explicit caching is enabled, false otherwise.
Note:

You enable implicit and explicit caching for a particular physical connection independently. Therefore, it is possible to do statement caching both implicitly and explicitly during the same session. Implicit and explicit statement caching share the same cache. Remember this when you set the statement cache size.

Disabling Explicit Statement Caching Disable explicit statement caching by calling setExplicitCachingEnabled(false). Disabling caching or closing the cache purges the cache. The following example disables explicit statement caching:
((OracleConnection)conn).setExplicitCachingEnabled(false);

Closing a Cached Statement

Perform the following to close a Satement and assure that it is not returned to the cache:

Disable caching for that statement

stmt.setDisableStatementCaching(true);

Call the close method of the statement object

stmt.close();

Physically Closing a Cached Statement With implicit statement caching enabled, you cannot physically close statements manually. The close method of a statement object caches the statement instead of closing it. The statement is physically closed automatically under one of following three conditions:

When the associated connection is closed When the cache reaches its size limit and the least recently used statement object is preempted from cache by the LRU scheme If you call the close method on a statement for which statement caching is disabled

Using Implicit Statement Caching

Once you enable implicit statement caching, by default, all prepared and callable statements are automatically cached. Implicit statement caching includes the following steps:
1. 2.

Enable implicit statement caching. Allocate a statement using one of the standard methods.

Statement and Resultset Caching

20-5

Using Statement Caching

3. 4. 5.

Disable implicit statement caching for any particular statement you do not want to cache. This is an optional step. Cache the statement using the close method. Retrieve the implicitly cached statement by calling the appropriate standard prepare method.

Allocating a Statement for Implicit Caching To allocate a statement for implicit statement caching, use either the prepareStatement or prepareCall method as you would typically. The following code allocates a new statement object called pstmt:
PreparedStatement pstmt = conn.prepareStatement ("UPDATE emp SET ename = ? WHERE rowid = ?");

Disabling Implicit Statement Caching for a Particular Statement With implicit statement caching enabled for a connection, by default, all callable and prepared statements of that connection are automatically cached. To prevent a particular callable or prepared statement from being implicitly cached, use the setDisableStatementCaching method of the statement object. You can manage cache space by calling the setDisableStatementCaching method on any infrequently used statement. The following code disables implicit statement caching for pstmt:
PreparedStatement pstmt = conn.prepareStatement("SELECT 1 from DUAL"); ((OraclePreparedStatement)pstmt).setDisableStmtCaching(true); pstmt.close ();

Note:

If you are using JSE 6, then you can disable Statement caching by using the standard JDBC 4.0 method setPoolable:

PreparedStatement.setPoolable(false);

Use the following to check whether the Statement object is poolable:

Statement.isPoolable();

Implicitly Caching a Statement To cache an allocated statement, call the close method of the statement object. When you call the close method on an OraclePreparedStatement or OracleCallableStatement object, the JDBC driver automatically puts this statement in cache, unless you have disabled caching for this statement. The following code caches the pstmt statement:
pstmt.close ();

Retrieving an Implicitly Cached Statement To retrieve an implicitly cached statement, call either the prepareStatement or prepareCall method, depending on the statement type. The following code retrieves pstmt from cache using the prepareStatement method:

20-6 Oracle Database JDBC Developer's Guide and Reference

Using Statement Caching

pstmt = conn.prepareStatement ("UPDATE emp SET ename = ? WHERE rowid = ?");

Table 202 describes the methods used to allocate statements and retrieve implicitly cached statements.
Table 202 Method prepareStatement Methods Used in Statement Allocation and Implicit Statement Caching Functionality for Implicit Statement Caching Performs a cache search that either finds and returns the desired cached OraclePreparedStatement object or allocates a new OraclePreparedStatement object if a match is not found Performs a cache search that either finds and returns the desired cached OracleCallableStatement object or allocates a new OracleCallableStatement object if a match is not found

prepareCall

Using Explicit Statement Caching

A prepared or callable statement can be explicitly cached when you enable explicit statement caching. Explicit statement caching includes the following steps:
1. 2. 3. 4. 5.

Enable explicit statement caching. Allocate a statement using one of the standard methods. Explicitly cache the statement by closing it with a key, using the closeWithKey method. Retrieve the explicitly cached statement by calling the appropriate Oracle WithKey method, specifying the appropriate key. Re-cache an open, explicitly cached statement by closing it again with the closeWithKey method. Each time a cached statement is closed, it is re-cached with its key.

Allocating a Statement for Explicit Caching To allocate a statement for explicit statement caching, use either the createStatement, prepareStatement, or prepareCall method as you would typically. The following code allocates a new statement object called pstmt:
PreparedStatement pstmt = conn.prepareStatement ("UPDATE emp SET ename = ? WHERE rowid = ?");

Explicitly Caching a Statement To explicitly cache an allocated statement, call the closeWithKey method of the statement object, specifying a key. The key is an arbitrary Java String that you provide. The closeWithKey method caches a statement as is. This means the data, state, and metadata are retained and not cleared. The following code caches the pstmt statement with the key "mykey":
((OraclePreparedStatement)pstmt).closeWithKey ("mykey");

Retrieving an Explicitly Cached Statement To recall an explicitly cached statement, call either the getStatementWithKey or getCallWithKey methods depending on the statement type.

Statement and Resultset Caching

20-7

Reusing Statements Objects

If you retrieve a statement with a specified key, then the JDBC driver searches the cache for the statement, based on the specified key. If a match is found, then the matching statement is returned along with its state, data, and metadata. This information is as it was when the statement was last closed. If a match is not found, then the JDBC driver returns null. The following code recalls pstmt from cache using the "mykey" key with the getStatementWithKey method. Recall that the pstmt statement object was cached with the "mykey" key.
pstmt = ((OracleConnection)conn).getStatementWithKey ("mykey");

If you call the creationState method on the pstmt statement object, then the method returns EXPLICIT.
Important:

When you retrieve an explicitly cached statement, ensure that you use the method that is appropriate for your statement type when specifying the key. For example, if you used the prepareStatement method to allocate a statement, then use the getStatementWithKey method to retrieve that statement from cache. The JDBC driver does not verify the type of statement it is returning.

Table 203 describes the methods used to retrieve explicitly cached statements.
Table 203 Method getStatementWithKey getCallWithKey Methods Used to Retrieve Explicitly Cached Statements Functionality for Explicit Statement Caching Specifies the key needed to retrieve a prepared statement from cache Specifies the key needed to retrieve a callable statement from cache

Reusing Statements Objects

The JDBC 3.0 specification introduces the feature of statement pooling that allows an application to reuse a PreparedStatement object in the same way as it uses a Connection object. The PreparedStatement objects can be reused by multiple logical connections in a transparent manner. This section covers the following topics:

Using a Pooled Statement Closing a Pooled Statement

Using a Pooled Statement

An application can find out whether a data source supports statement pooling by calling the isPoolable method from the Statement interface. If the return value is true, then the application knows that the PreparedStatement object is being pooled. The application can also request a statement to be pooled or not pooled by using the setPoolable method from the Statement interface. Reusing of pooled statement should be completely transparent to the application, that is, the application code should remain the same whether a PreparedStatement object participates in statement pooling or not. If an application closes a

20-8 Oracle Database JDBC Developer's Guide and Reference

Resultset Caching

PreparedStatement object, it must still call Connection.prepareStatement method in order to reuse it.
Note:

An application has no direct control over how statements are pooled. A pool of statements is associated with a PooledConnection object, whose behavior is determined by the properties of the ConnectionPoolDataSource object that produced it.

Closing a Pooled Statement

An application closes a pooled statement exactly the same way it closes a nonpooled statement. Once a statement is closed, whether is it pooled or nonpooled, it is no longer available for use by the application and an attempt to reuse it causes an exception to be thrown. The only difference visible is that an application cannot directly close a physical statement that is being pooled. This is done by the pool manager. The method PooledConnection.closeAll closes all of the statements open on a given physical connection, which releases the resources associated with those statements. The following methods can close a pooled statement:

close This java.sql.Statement interface method is called by an application. If the statement is being pooled, then it closes the logical statement used by the application but does not close the physical statement being pooled

close This java.sql.Connection interface method is called by an application. This method acts differently depending upon whether the connection using the statement is being pooled or not: Nonpooled connection This method closes the physical connection and all statements created by that connection. This is necessary because the garbage collection mechanism is unable to detect when externally managed resources can be released. Pooled connection This method closes the logical connection and the logical statements it returned, but leaves open the underlying PooledConnection object and any associated pooled statements

PooledConnection.closeAll This method is called by the connection pool manager to close all of the physical statements being pooled by the PooledConnection object

Resultset Caching
Your applications sometime require to retrieve huge data from the database, using JDBC, in operations like searching data. When you request for a search, the application might return the whole result set at once. This process takes lot of time and has an impact on performance. This problem can be solved in the following ways:

Server-side Cache Client Query Cache

Statement and Resultset Caching 20-9

Resultset Caching

Server-side Cache
Oracle Database 11g Release 1 (11.1) provides support for server-side resultset caching for both JDBC types. The server-side result cache is used to cache the results of the current query or query fragment in memory and then to use the cached results in future executions of the query or query fragment. The cached results reside in the result cache memory portion of the SGA. A cached result is automatically invalidated whenever a database object used in its creation is successfully modified.
See Also: Oracle Database Performance Tuning Guide for information about using server-side cache

Client Query Cache

Oracle Database 11g Release 1 (11.1) provides support for client result cache for JDBC OCI client. The client query cache improves performance of applications by caching query result sets in a way that subsequent query executions can access the cached result set without fetching rows from the server. This eliminates some round-trips to the server and reduces CPU usage on the server. This allows significant improvements in response time for frequent client SQL query executions and for fetching rows. The scalability on the server is increased since it expends less CPU time.
See Also: Oracle Call Interface Programmer's Guide for more information about client query cache.

20-10 Oracle Database JDBC Developer's Guide and Reference

21
Implicit Connection Caching
Connection caching, generally implemented in the middle tier, is a means of keeping and using cache of physical database connections.
Note:

The previous cache architecture, based on OracleConnectionCache and OracleConnectionCacheImpl, is deprecated. We recommend that you take advantage of the new architecture, which is more powerful and offers better performance.

The implicit connection cache is an improved Java Database Connectivity (JDBC) 3.0-compliant connection cache implementation for DataSource. Java and Java2 Platform, Enterprise Edition (J2EE) applications benefit from transparent access to the cache, support for multiple users, and the ability to request connections based on user-defined profiles. An application turns the implicit connection cache on by calling setConnectionCachingEnabled(true) on an OracleDataSource. After implicit caching is turned on, the first connection request to the OracleDataSource transparently creates a connection cache. There is no need for application developers to write their own cache implementations. This chapter is divided into the following sections:

The Implicit Connection Cache Using the Connection Cache Connection Attributes Connection Cache Properties Connection Cache Manager API Advanced Topics
Note:

The concept of connection caching is not relevant to the server-side internal driver, where you always use the default connection. Connection caching is only relevant to the client-side JDBC drivers and the server-side Thin driver.

Implicit Connection Caching

21-1

The Implicit Connection Cache

The Implicit Connection Cache

The connection caching architecture has been redesigned so that caching is transparently integrated into the data source architecture. The connection cache uses the concept of physical connections and logical connections. Physical connections are the actual connections returned by the database and logical connections are containers used by the cache to manipulate physical connections. You can think of logical connections as handles. The caches always return logical connections, which implement the same interfaces as physical connections. The implicit connection cache offers:

Driver independence Both the JDBC Thin and JDBC Oracle Call Interface (OCI) drivers support the implicit connection cache.

Transparent access to the JDBC connection cache After an application turns implicit caching on, it uses the standard OracleDataSource application programming interfaces (APIs) to get connections. With caching enabled, all connection requests are serviced from the connection cache. When an application calls OracleConnection.close to close the logical connection, the physical connection is returned to the cache.

Single cache per OracleDataSource instance When connection caching is turned on, each cache enabled OracleDataSource has exactly one cache associated with it. All connections obtained through that data source, no matter what user name and password are used, are returned to the cache. When an application requests a connection from the data source, the cache either returns an existing connection or creates a new connection with matching authentication information.
Note: Caches cannot be shared between DataSource instances. There is a one-to-one mapping between cache enabled DataSource instances and caches.

Heterogeneous user names and passwords per cache Unlike in the previous cache implementation, all connections obtained through the same data source are stored in a common cache, no matter what user name and password the connection requests.

Support for JDBC 3.0 connection caching, including support for multiple users and the required cache properties. Property-based configuration Cache properties define the behavior of the cache. The supported properties set timeouts, the number of connections to be held in the cache, and so on. Using these properties, applications can reclaim and reuse abandoned connections. The implicit connection cache supports all the JDBC 3.0 connection cache properties.

OracleConnectionCacheManager The new OracleConnectionCacheManager class provides a rich set of administrative APIs that applications can use to manage the connection cache. Each virtual machine has one distinguished instance of

21-2 Oracle Database JDBC Developer's Guide and Reference

Using the Connection Cache

OracleConnectionCacheManager. Applications manage a cache through the single OracleConnectionCacheManager instance.

User-defined connection attributes The implicit connection cache supports user-defined connection attributes that can be used to determine which connections are retrieved from the cache. Connection attributes can be thought of as labels whose semantics are defined by the application, not by the caching mechanism.

Callback mechanism The implicit connection cache provides a mechanism for users to define cache behavior when a connection is returned to the cache, when handling abandoned connections, and when a connection is requested but none is available in the cache.

Connect-time load balancing Implicit connection caching provides connect time load balancing when a connection is first created by the application. The database listener distributes connection creation across Oracle Real Application Cluster instances that would perform the best at the time of connection creation.
See Also: Oracle Real Application Clusters Administration and Deployment Guide

Run-time connection load balancing Run-time connection load balancing of work requests uses Service Metrics to route work requests to an Oracle Real Application Cluster instance that offers the best performance. Selecting a connection from the cache based on service, to execute a work request, greatly increases the throughput and scalability.
See Also: Chapter 22, "Run-Time Connection Load Balancing" and Oracle Real Application Clusters Administration and Deployment Guide

Using the Connection Cache

This section discusses how applications use the implicit connection cache. It covers the following topics:

Turning Caching On Opening a Connection Setting Connection Cache Name Setting Connection Cache Properties Closing A Connection Implicit Connection Cache Example

Turning Caching On
An application turns the implicit connection cache on by calling OracleDataSource.setConnectionCachingEnabled(true). After implicit caching is turned on, the first connection request to the OracleDataSource transparently creates a connection cache. Example 211 provides a sample code that uses the implicit connection cache.

Implicit Connection Caching

21-3

Using the Connection Cache

Example 211

Using the Implicit Connection Cache

// Example to show binding of OracleDataSource to JNDI, // then using implicit connection cache import oracle.jdbc.pool.*; // import the pool package Context ctx = new InitialContext(ht); OracleDataSource ods = new OracleDataSource(); // Set DataSource properties ods.setUser("Scott"); ods.setConnectionCachingEnabled(true); // Turns on caching ctx.bind("MyDS", ods); // ... // Retrieve DataSource from the InitialContext ods =(OracleDataSource) ctx. lookup("MyDS"); // Transparently create cache and retrieve connection conn = ods.getConnection(); // ... conn.close(); // return connection to the cache // ... ods.close() // close datasource and clean up the cache

Opening a Connection
After you have turned connection caching on, whenever you retrieve a connection through OracleDataSource.getConnection, the JDBC drivers check to see if a connection is available in the cache. The getConnection method checks if there are any free physical connections in the cache that match the specified criteria. If a match is found, then a logical connection is returned wrapping the physical connection. If no physical connection match is found, then a new physical connection is created, wrapped in a logical connection, and returned. There are four variations on getConnection, two that make no reference to the connection cache and two that specify which sort of connections the cache may return. The non-cache-specific getConnection methods behave as standard. When implicit connection cache is enabled, the connection returned by OracleDataSource.getConnection may not have the state reset. You must, therefore, reset all the connection states, such as auto-commit, batch size, prefetch size, transaction status, and transaction isolation, before putting the connection back into the cache.
Note:

Setting Connection Cache Name

The ConnectionCacheName property of OracleDataSource is an optional property used by the Connection Cache manager to manage a connection cache. You can set this property by calling the following method:
public void synchronized setConnectionCacheName(String cacheName) throws SQLException

21-4 Oracle Database JDBC Developer's Guide and Reference

Using the Connection Cache

When this property is set, the name is used to uniquely identify the cache accessed by the cache-enabled OracleDataSource. If the property is not set, then a default cache name is created using the convention DataSourceName#HexRepresentationOfNumberOfCaches.
Note: The getConnectionCacheName() method will return the name of the connection cache only if the setConnectionCacheName method is called after the setConnectionCachingEnabled method is called.

Setting Connection Cache Properties

You can fine-tune the behavior of the implicit connection cache using the setConnectionCacheProperties method to set various connection properties.
See Also:

"Connection Cache Properties" on page 21-8

Note:

Before setting the cache-specific properties, you need to enable caching on the data source; otherwise the setConnectionCacheProperties method will have no effect. Although these properties govern the behavior of the connection cache, they are set on the data source, and not on the connection or on the cache itself.

Closing A Connection
An application returns a connection to the cache by calling close. There are two variants of the close method: one with no arguments and one that takes a Connection object as argument.
Note:

Applications must close connections to ensure that the connections are returned to the cache. When implicit connection cache is enabled, you must reset all the connection states, such as auto-commit, batch size, prefetch size, transaction status, and transaction isolation, before putting the connection back into the cache. This ensures that any subsequent connection retrieved from the cache will have its state reset.

Implicit Connection Cache Example

Example 212 demonstrates creating a data source, setting its caching and data source properties, retrieving a connection, and closing that connection in order to return it to the cache.
Example 212 Connection Cache Example

import java.sql.*; import javax.sql.*; import java.util.*;

Implicit Connection Caching

21-5

Connection Attributes

import import import import ...

javax.naming.*; javax.naming.spi.*; oracle.jdbc.*; oracle.jdbc.pool.*;

// create a DataSource OracleDataSource ods = new OracleDataSource(); // set cache properties java.util.Properties prop = new java.util.Properties(); prop.setProperty("MinLimit", "2"); prop.setProperty("MaxLimit", "10"); // set DataSource properties String url = "jdbc:oracle:oci8:@"; ods.setURL(url); ods.setUser("hr"); ods.setPassword("hr"); ods.setConnectionCachingEnabled(true); // be sure set to true ods.setConnectionCacheProperties (prop); ods.setConnectionCacheName("ImplicitCache01"); // this cache's name // We need to create a connection to create the cache Connection conn = ds.getConnection(user, pass); Statement stmt = conn1.createStatement(); ResultSet rset = stmt.executeQuery("select user from dual"); conn1.close(); ods.close();

Connection Attributes
Each connection obtained from a data source can have user-defined attributes. Attributes are specified by the application developer and are java.lang.Properties name/value pairs. An application can use connection attributes to supply additional semantics to identify connections. For instance, an application may create an attribute named connection_type and then assign it the value payroll or inventory.
Note:

The semantics of connection attributes are entirely application-defined. The connection cache itself enforces no restrictions on the key or value of connection attributes.

The methods that get and set connection attributes are found on OracleConnection. This section covers the following topics:

Getting Connections Setting Connection Attributes Checking Attributes of a Returned Connection Connection Attribute Example

21-6 Oracle Database JDBC Developer's Guide and Reference

Connection Attributes

Getting Connections
The first connection you retrieve has no attributes. You must set them. After you have set attributes on a connection, you can request the connection by specifying its attribute, using the specialized forms of getConnection:

getConnection(java.util.Properties cachedConnectionAttributes Requests a database connection that matches the specified cachedConnectionAttributes.
See Also:

For a discussion on connection attributes, see "Other Properties" on page 21-10.

getConnection(java.lang.String user, java.lang.String password, java.util.Properties cachedConnectionAttributes) Requests a database connection from the implicit connection cache that matches the specified user, password and cachedConnectionAttributes. If null values are passed for user and password, the DataSource defaults are used.

Attribute Matching Rules The rules for matching connectionAttributes come in two variations:

Basic In this, the cache is searched to retrieve the connection that matches the attributes. The connection search mechanism as follows:
1. 2.

If an exact match is found, the connection is returned to the caller. If an exact match is not found and the ClosestConnectionMatch data source property is set, then the connection with the closest match is returned. The closest matched connection is one that has the highest number of the original attributes matched. Note that the closest matched connection may match a subset of the original attributes, but does not have any attributes that are not part of the original list. For example, if the original list of attributes is A, B and C, then a closest match may have A and B set, but never a D. If none of the existing connections match, a new connection is returned. The new connection is created using the user name and password set on the DataSource. If getConnection(String, String, java.util.Properties) is called, then the user name and password passed as arguments are used to open the new connection.

3.

Advanced, where attributes may be associated with weights. The connection search mechanism is similar to the basic connectionAttributes based search, except that the connections are searched not only based on the connectionAttributes, but also using a set of weights that are associated with the keys on the connectionAttributes. These weights are assigned to the keys as a one time operation and is supported as a connection cache property, AttributeWeights.

Setting Connection Attributes

An application sets connection attributes using:
applyConnectionAttributes(java.util.Properties connAttr)

Implicit Connection Caching

21-7

Connection Cache Properties

No validation is done on connAttr. Applying connection attributes is cumulative. Each time you call applyConnectionAttributes, the connAttr attribute you supply is added to those previously in force.

Checking Attributes of a Returned Connection

When an application requests a connection with specified attributes, it is possible that no match will be found in the connection cache. When this happens, the connection cache creates a connection with no attributes and returns it. The connection cache cannot create a connection with the requested attributes, because the cache manager is ignorant of the semantics of the attributes.
Note:

If the closestConnectionMatch property has been set, then the cache manager looks for close attribute matches rather than exact matches.

For this reason, applications should always check the attributes of a returned connection. To do this, use the getUnMatchedConnectionAttributes method, which returns a list of any attributes that were not matched in retrieving the connection. If the return value of this method is null, you know that you must set all the connection attributes.

Connection Attribute Example

Example 213 illustrates using connection attributes.
Example 213 Using Connection Attributes

java.util.Properties connAttr = new java.util.Properties(); connAttr.setProperty("connection_type", "payroll"); // retrieve connection that matches attributes Connection conn = ds.getConnection(connAttr); // Check to see which attributes weren't matched unmatchedProp = ((OracleConnection)conn).getUnMatchedConnectionAttributes(); if ( unmatchedProp != null ) { // apply attributes to the connection ((OracleConnection)conn).applyConnectionAttributes(connAttr); } // verify whether conn contains property after apply attributes connProp = ((OracleConnection)conn).getConnectionAttributes(); listProperties (connProp);

Connection Cache Properties

The connection cache properties govern the characteristics of a connection cache. This section lists the supported connection cache properties. It covers the following topics:

Limit Properties TIMEOUT Properties Other Properties Connection Property Example

21-8 Oracle Database JDBC Developer's Guide and Reference

Connection Cache Properties

Applications set cache properties in one of the following ways:

Using the OracleDataSource method setConnectionCacheProperties When creating a cache using OracleConnectionCacheManager When re-initializing a cache using OracleConnectionCacheManager

Limit Properties
These properties control the size of the cache. InitialLimit Sets how many connections are created in the cache when it is created or reinitialized. When this property is set to an integer value greater than 0, creating or reinitializing the cache automatically creates the specified number of connections, filling the cache in advance of need. Default: 0 MaxLimit Sets the maximum number of connection instances the cache can hold. The default value is Integer.MAX_VALUE, meaning that there is no limit enforced by the connection cache, so that the number of connections is limited only by the number of database sessions configured for the database. Default: Integer.MAX_VALUE (no limit) MaxStatementsLimit Sets the maximum number of statements that a connection keeps open. When a cache has this property set, reinitializing the cache or closing the data source automatically closes all cursors beyond the specified MaxStatementsLimit. Default: 0 MinLimit Sets the minimum number of connections the cache maintains. Default: 0
Note:

Setting the MinLimit property does not initialize the cache to contain the minimum number of connections. To do this, use the InitialLimit property. When InitialLimit is greater than MinLimit, it is possible to have any number of connections specified by InitialLimit up to a value specified by MaxLimit. Therefore, InitialLimit does not depend on MinLimit. Connections can fall below the minimum limit set on the connection pool when JDBC Fast Connection Failover DOWN events are processed. The processing removes affected connections from the pool. MinLimit will be honored as requests to the connection pool ramp up and the number of connections get past the MinLimit value.

Implicit Connection Caching

21-9

Connection Cache Properties

TIMEOUT Properties
These properties control the lifetime of an element in the cache. InactivityTimeout Sets the maximum time a physical connection can remain idle in a connection cache. An idle connection is one that is not active and does not have a logical handle associated with it. When InactivityTimeout expires, the underlying physical connection is closed. However, the size of the cache is not allowed to shrink below minLimit, if it has been set. Default: 0 (no timeout in effect) TimeToLiveTimeout Sets the maximum time in seconds that a logical connection can remain open. When TimeToLiveTimeout expires, the logical connection is unconditionally closed, the relevant statement handles are canceled, and the underlying physical connection is returned to the cache for reuse. Default: 0 (no timeout in effect)
See Also: "Use Cases for TimeToLiveTimeout and AbandonedConnectionTimeout" on page 21-14

AbandonedConnectionTimeout Sets the maximum time that a connection can remain unused before the connection is closed and returned to the cache. A connection is considered unused if it has not had SQL database activity. When AbandonedConnectionTimeout is set, JDBC monitors SQL database activity on each logical connection. For example, when stmt.execute is called on the connection, a heartbeat is registered to convey that this connection is active. The heartbeats are set at each database execution. If a connection has been inactive for the specified amount of time, the underlying connection is reclaimed and returned to the cache for reuse. Default: 0 (no timeout in effect)
See Also: "Use Cases for TimeToLiveTimeout and AbandonedConnectionTimeout" on page 21-14

PropertyCheckInterval Sets the time interval at which the cache manager inspects and enforces all specified cache properties. PropertyCheckInterval is set in seconds. Default: 900 seconds

Other Properties
These properties control miscellaneous cache behavior. AttributeWeights AttributeWeights sets the weight for each attribute set on the connection.
See Also:

"AttributeWeights" on page 21-13

21-10 Oracle Database JDBC Developer's Guide and Reference

Connection Cache Properties

ClosestConnectionMatch ClosestConnectionMatch causes the connection cache to retrieve the connection with the closest approximation to the specified connection attributes.
See Also:

"ClosestConnectionMatch" on page 21-13

ConnectionWaitTimeout Specifies cache behavior when a connection is requested and there are already MaxLimit connections active. If ConnectionWaitTimeout is equal to zero, then each connection request waits for zero seconds, that is, null connection is returned immediately. If ConnectionWaitTimeout is greater than zero, then each connection request waits for the specified number of seconds or until a connection is returned to the cache. If no connection is returned to the cache before the timeout elapses, then the connection request returns null. Default: 0 LowerThresholdLimit Sets the lower threshold limit on the cache. The default is 20 percent of the MaxLimit on the connection cache. This property is used whenever a releaseConnection() cache callback method is registered. ValidateConnection Setting ValidateConnection to true causes the connection cache to test every connection it retrieves against the underlying database. If a valid connection cannot be retrieved, then an exception is thrown. Default: false

Connection Property Example

Example 214 demonstrates how an application uses connection properties.
Example 214 Using Connection Properties

import java.sql.*; import javax.sql.*; import java.util.*; import javax.naming.*; import javax.naming.spi.*; import oracle.jdbc.*; import oracle.jdbc.pool.*; ... OracleDataSource ds = (OracleDataSource) ctx.lookup("..."); java.util.Properties prop = new java.util.Properties (); prop.setProperty("MinLimit", "5"); // the cache size is 5 at least prop.setProperty("MaxLimit", "25"); prop.setProperty("InitialLimit", "3"); // create 3 connections at startup prop.setProperty("InactivityTimeout", "1800"); // seconds prop.setProperty("AbandonedConnectionTimeout", "900"); // seconds prop.setProperty("MaxStatementsLimit", "10"); prop.setProperty("PropertyCheckInterval", "60"); // seconds ds.setConnectionCacheProperties (prop); // set properties Connection conn = ds.getConnection(); conn.dosomework(); java.util.Properties propList=ds.getConnectionCacheProperties();

// retrieve

Implicit Connection Caching 21-11

Connection Cache Manager API

Connection Cache Manager API

OracleConnectionCacheManager provides administrative APIs that the middle tier can use to manage available connection caches. This section provides an example of using the Connection Cache Manager. Example Of ConnectionCacheManager Use Example 215 demonstrates the OracleConnectionCacheManager interfaces.
Example 215 Connection Cache Manager Example

import java.sql.*; import javax.sql.*; import java.util.*; import javax.naming.*; import javax.naming.spi.*; import oracle.jdbc.*; import oracle.jdbc.pool.*; ... // Get singleton ConnectionCacheManager instance OracleConnectionCacheManager occm = OracleConnectionCacheManager.getConnectionCacheManagerInstance(); String cacheName = "foo"; // Look for a specific cache // Use Cache Manager to check # of available connections // and active connections System.out.println(occm.getNumberOfAvailableConnections(cacheName) " connections are available in cache " + cacheName); System.out.println(occm.getNumberOfActiveConnections(cacheName) + " connections are active"); // Refresh all connections in cache occm.refreshCache(cacheName, OracleConnectionCacheManager.REFRESH_ALL_CONNECTIONS); // Reinitialize cache, closing all connections java.util.Properties newProp = new java.util.Properties(); newProp.setProperty("MaxLimit", "50"); occm.reinitializeCache(cacheName, newProp);

Advanced Topics
This section discusses cache functionality that is useful for advanced users, but is not essential to understanding or using the implicit connection cache. This section covers the following topics:

Attribute Weights And Connection Matching Connection Cache Callbacks

Attribute Weights And Connection Matching

There are two connection cache properties that enable the developer to specify which connections in the connection cache are accepted in response to a getConnection request. When you set the ClosestConnectionMatch property to true, you are telling the connection cache manager to return connections that match only some of the attributes you have specified. If you do not specify attributeWeights, then the connection cache manager returns the connection that matches the highest number of attributes. If you specify

21-12 Oracle Database JDBC Developer's Guide and Reference

Advanced Topics

attributeWeights, then you can control the priority the manager uses in matching attributes. ClosestConnectionMatch Setting ClosestConnectionMatch to true causes the connection cache to retrieve the connection with the closest approximation to the specified connection attributes. This can be used in combination with AttributeWeights to specify what is considered a closest match. Default: false AttributeWeights Sets the weights for each connectionAttribute. Used when ClosestConnectionMatch is set to true to determine which attributes are given highest priority when searching for matches. An attribute with a high weight is given more importance in determining a match than an attribute with a low weight. AttributeWeights contains a set of Strings representing key/value pairs. Each key/value pair sets the weights for each connectionAttribute for which the user intends to request a connection. Each String is in the format written by the java.util.Properties.Store(OutputStream, String) method, and thus can be read by the java.util.Properties.load(InputStream) method. The Key is a connectionAttribute and the Value is the weight. A weight must be an integer value greater than 0. The default weight is 1. For example, TRANSACTION_ISOLATION could be assigned a weight of 10 and ROLE a weight of 5. If ClosestConnectionMatch is set to true, when a connectionAttribute based connection request is made on the cache, connections with a matching TRANSACTION_ISOLATION will be favored over connections with a matching ROLE. Default: No AttributeWeights

Connection Cache Callbacks

The implicit connection cache offers a way for the application to specify callbacks to be called by the connection cache. Callback methods are supported with the OracleConnectionCacheCallback interface. This callback mechanism is useful to take advantage of the special knowledge of the application about particular connections, supplementing the default behavior when handling abandoned connections or when the cache is empty. OracleConnectionCacheCallback is an interface that must be implemented by the user and registered with OracleConnection. The registration API is:
public void registerConnectionCacheCallback( OracleConnectionCacheCallback cbk, Object usrObj,

int cbkflag);

In this interface, cbk is the user implementation of the OracleConnectionCacheCallback interface. The usrObj parameter contains any parameters that the user wants supplied. This user object is passed back, unmodified, when the callback method is called. The cbkflag parameter specifies which callback method should be called. It must be one of the following values:

OracleConnection.ABANDONED_CONNECTION_CALLBACK OracleConnection.RELEASE_CONNECTION_CALLBACK

Implicit Connection Caching 21-13

Advanced Topics

OracleConnection.ALL_CALLBACKS

When ALL_CALLBACKS is set, all the connection cache callback methods are called. For example,
// register callback, to invoke all callback methods ((OracleConnection)conn).registerConnectionCacheCallback( new UserConnectionCacheCallback(), new SomeUserObject(), OracleConnection.ALL_CALLBACKS);

An application can register a ConnectionCacheCallback on an OracleConnection. When a callback is registered, the connection cache calls the handleAbandonedConnection method of the callback before reclaiming the connection. If the callback returns true, then the connection is reclaimed. If the callback returns false, then the connection remains active. The UserConnectionCacheCallback interface supports two callback methods to be implemented by the user, releaseConnection and handleAbandonedConnection.

Use Cases for TimeToLiveTimeout and AbandonedConnectionTimeout

The following are the use cases for the TimeToLiveTimeout and AbandonedConnectionTimeout timeout mechanisms when used with implicit connection cache. Note that these timeout mechanisms are applicable to the logical connection when it is retrieved from the connection cache.

The application considers the connections are completely stateless. When the connections are stateless either of the timeout mechanisms can be used. The connections for which the timeout expires are put back into the connection cache for reuse. These connections are valid for reuse because there is no session state associated with them.

The application maintains state on each connection, but has a cleanup routine that can render the connections stateless when they are returned to the connection cache. In this case, TimeToLiveTimeout cannot be used. There is no way for the connection cache to ensure that a connection returned to the cache is in a reusable condition. However, AbandonedConnectionTimeout can be used in this scenario, only if OracleConnectionCacheCallback is registered on the connection. The handleAbandonedConnection callback method is implemented by the application and ensures that the necessary cleanup is done. The connection is closed when the timeout processing invokes this callback method. The closing of this connection by the call back method causes the connection to be put back into the connection cache in a state where it is reusable.
Note:

Do not to close the connection after calling handleAbandonedConnection method because the connection could be in an invalid state. JDBC internally knows how to reclaim a connection even when it is in an invalid state. "Connection Cache Callbacks" on page 21-13

See Also:

21-14 Oracle Database JDBC Developer's Guide and Reference

Advanced Topics

The application maintains state on each connection, but has no control over the connection and, therefore, cannot ensure cleaning up of state for reuse of connections by other applications or users. The use of either of the timeout mechanism is not recommended.

Implicit Connection Caching 21-15

Advanced Topics

21-16 Oracle Database JDBC Developer's Guide and Reference

22
Run-Time Connection Load Balancing
Oracle Database 10g provides the run-time connection load balancing feature. This chapter contains the following sections:

Overview Run-Time Connection Load Balancing Enabling Run-Time Connection Load Balancing

Overview
In an Oracle Real Application Clusters environment, a connection could belong to any instance that provides the relevant service. In the best case, all instances perform equally well and randomly retrieving a connection from the cache is appropriate. However, when one instance performs better than others, random selection of a connection is inefficient. The run-time connection load balancing feature enables routing of work requests to an instance that offers the best performance, minimizing the need to relocate work.

Run-Time Connection Load Balancing

Figure 221 illustrates run-time connection load balancing. When run-time connection load balancing is enabled on the implicit connection cache, the following steps occur:
1. 2.

A client requests a connection from the connection cache by calling the getConnection method on the DataSource object. The run-time connection load balancing mechanism selects the connection that belongs to the best instance from the connection cache. In Figure 221, this could be either Instance1 or Instance2. The client receives the connection that would process the work request with the best response time.

3.

Run-Time Connection Load Balancing 22-1

Enabling Run-Time Connection Load Balancing

Figure 221 Run-Time Connection Load Balancing

Implicit Connection Cache

RAC Database

2 Instance1Best

Client 1 Instance2Best

Instance3Bad

Connection retrieval based on the Load Balancing Advisory is automatic. A request for a connection is serviced by selecting a connection based on the service goal as determined by the Load Balancing Advisory. The service goal determines whether the connection provides best service quality, that is, how efficiently a single transaction completes, or best throughput, that is, how efficiently an entire job or long-running query completes. The advisory is used by the connection cache as long as the events are posted by Oracle Real Application Clusters. When the events stop arriving, the connection cache reverts to random retrieval of connections from the cache. Run-time connection load balancing relies on the Oracle Notification Service (ONS) infrastructure. It uses the same out-of-band ONS event mechanism that is used for Fast Connection Failover processing. As a result, run-time connection load balancing is enabled by default when Fast Connection Failover is enabled. There is no additional setup or configuration of ONS required to benefit from run-time connection load balancing.
See Also:

"Using Fast Connection Failover" on page 26-2

Enabling Run-Time Connection Load Balancing

To enable and use run-time connection load balancing, you must configure the Oracle Real Application Clusters Database in the following manner:

The service goal must be set to one of the following: DBMS_SERVICE.SERVICE_TIME DBMS_SERVICE.THROUGHPUT

The connection balancing goal must be set to SHORT.

These goals need to be set when calling dbms_service.create_service or dbms_service.modify_service. The service goal can be set using the goal parameter, and the connection balancing goal can be set using the clb_goal parameter.

22-2 Oracle Database JDBC Developer's Guide and Reference

Enabling Run-Time Connection Load Balancing

Note: You can set the connection balancing goal to LONG. However, this is mostly useful for closed workloads, that is, when the rate of completing work is equal to the rate of starting new work.

See Also: Oracle Real Application Clusters Administration and Deployment Guide

Run-Time Connection Load Balancing 22-3

Enabling Run-Time Connection Load Balancing

22-4 Oracle Database JDBC Developer's Guide and Reference

23
Performance Extensions
This chapter describes the Oracle performance extensions to the Java Database Connectivity (JDBC) standard. This chapter covers the following topics:

Update Batching Additional Oracle Performance Extensions

Update Batching
You can reduce the number of round-trips to the database, thereby improving application performance, by grouping multiple UPDATE, DELETE, or INSERT statements into a single batch and having the whole batch sent to the database and processed in one trip. This is referred to as update batching.
Note:

The JDBC 2.0 specification refers to update batching as batch updates.

This is especially useful with prepared statements, when you are repeating the same statement with different bind variables. Oracle JDBC supports two distinct models for update batching:

The standard model, implementing the JDBC 2.0 specification, which is referred to as standard update batching The Oracle-specific model, independent of the JDBC 2.0 specification, which is referred to as Oracle update batching
Note:

It is important to be aware that you cannot mix these models. In any single application, you can use one model or the other, but not both. Oracle JDBC driver will throw exceptions when you mix these.

This section covers the following topics:

Overview of Update Batching Models Oracle Update Batching Standard Update Batching Premature Batch Flush
Performance Extensions 23-1

Update Batching

Overview of Update Batching Models

This section compares and contrasts the general models and types of statements supported for standard update batching and Oracle update batching. Oracle Model versus Standard Model Oracle update batching uses a batch value that typically results in implicit processing of a batch. The batch value is the number of operations you want to add to a batch for each trip to the database. As soon as that many operations have been added to the batch, the batch is processed. Note the following:

You can set a default batch for the connection object, which applies to any prepared statement run in that connection. For any individual prepared statement object, you can set a statement batch value that overrides the connection batch value. You can choose to explicitly process a batch at any time, overriding both the connection batch value and the statement batch value.

Standard update batching is a manual, explicit model. There is no batch value. You manually add operations to the batch, and then, explicitly choose when to process the batch. Oracle update batching is a more efficient model because the driver knows ahead of time how many operations will be batched. In this sense, the Oracle model is more static and predictable. With the standard model, the driver has no way of knowing in advance how many operations will be batched. In this sense, the standard model is more dynamic in nature. If you want to use update batching, then you can choose between the two models on the basis of the following:

Use Oracle update batching if portability is not critical. This will probably result in the greatest performance improvement. Use standard update batching if portability is a higher priority than performance.

Types of Statements Supported As implemented by Oracle, update batching is intended for use with prepared statements, when you are repeating the same statement with different bind variables. Be aware of the following:

Oracle update batching supports only prepared statement objects. For a callable statement, both the connection default batch value and the statement batch value are overridden with a value of 1. In an Oracle generic statement, there is no statement batch value, and the connection default batch value is overridden with a value of 1. To adhere to the JDBC 2.0 standard, Oracle implementation of standard update batching supports callable statements, without OUT parameters, and generic statements, as well as prepared statements. You can migrate standard update batching into an Oracle JDBC application without difficulty. You can batch only UPDATE, INSERT, or DELETE operations. Processing a batch that includes an operation that attempts to return a result set will cause an exception.

23-2 Oracle Database JDBC Developer's Guide and Reference

Update Batching

Note: The Oracle implementation of standard update batching does not implement true batching for generic statements and callable statements. Although Oracle JDBC supports the use of standard batching syntax for Statement and CallableStatement objects, you will see performance improvement for only PreparedStatement objects.

Oracle Update Batching

The Oracle update batching feature associates a batch value with each prepared statement object. With Oracle update batching, instead of the JDBC driver running a prepared statement each time executeUpdate method is called, the driver adds the statement to a batch of accumulated processing requests. The driver will pass all the operations to the database for processing once the batch value is reached. For example, if the batch value is 10, then each batch of 10 operations will be sent to the database and processed in one trip. A method in the OracleConnection class enables you to set a default batch value for the Oracle connection as a whole, and this batch value applies to any Oracle prepared statement in the connection. For any particular Oracle prepared statement, a method in the OraclePreparedStatement class enables you to set a statement batch value that overrides the connection batch value. You can also override both batch values by choosing to manually process the pending batch.
Note:

Do not mix standard update batching with Oracle update batching in the same application. The JDBC driver will throw an exception when you mix these. Disable auto-commit mode if you use either update batching model. In case an error occurs while you are processing a batch, this provides you the option of committing or rolling back the operations that ran successfully prior to the error.

Oracle Update Batching Characteristics and Limitations

Note the following limitations and implementation details regarding Oracle update batching:

By default, there is no statement batch value and the connection batch value is 1. Batch values between 5 and 30 tend to be the most effective. Setting a very high value might even have a negative effect. It is worth trying different values to verify the effectiveness for your particular application. Regardless of the batch value in effect, if any of the bind variables of an Oracle prepared statement is a stream type, then Oracle JDBC driver sets the batch value to 1 and sends any queued requests to the database for processing. Oracle JDBC driver automatically runs the sendBatch method of an Oracle prepared statement in any of the following circumstances: The connection receives a COMMIT request, either as a result of calling the commit method or as a result of auto-commit mode. The statement receives a close request. The connection receives a close request.
Performance Extensions 23-3

Update Batching

Note:

A connection COMMIT request, statement close, or connection close has an effect on a pending batch only if you use Oracle update batching. However, if you use standard update batching, then it has no effect on a pending batch.

If the connection receives a ROLLBACK request before sendBatch has been called, then the pending batched operations are not removed. You must explicitly call clearBatch to do this.

Setting the Connection Batch Value

You can specify a default batch value for any Oracle prepared statement in your Oracle connection. To do this, use the setDefaultExecuteBatch method of the OracleConnection object. For example, the following code sets the default batch value to 20 for all prepared statement objects associated with the conn connection object:
((OracleConnection)conn).setDefaultExecuteBatch(20);

Even though this sets the default batch value for all the prepared statements of the connection, you can override it by calling the setExecuteBatch method of the oracle.jdbc.OraclePreparedStatement interface on individual Oracle prepared statements. The connection batch value will apply to statement objects created after this batch value was set. Note that instead of calling setDefaultExecuteBatch method, you can set the defaultBatchValue Java property if you use a Java Properties object in establishing the connection.

Setting the Statement Batch Value

Use the following steps to set the statement batch value for a particular Oracle prepared statement. This will override any connection batch value set using the setDefaultExecuteBatch method of the OracleConnection instance for the connection in which the statement is processed.
1.

Write your prepared statement, and specify input values for the first row, as follows:
PreparedStatement ps = conn.prepareStatement ("INSERT INTO dept VALUES (?,?,?)"); ps.setInt (1,12); ps.setString (2,"Oracle"); ps.setString (3,"USA");

2.

Cast your prepared statement to OraclePreparedStatement, and apply the setExecuteBatch method. In this example, the batch size of the statement is set to 2.
((OraclePreparedStatement)ps).setExecuteBatch(2);

If you wish, insert the getExecuteBatch method at any point in the program to check the default batch value for the statement, as follows:
System.out.println (" Statement Execute Batch Value " + ((OraclePreparedStatement)ps).getExecuteBatch());

23-4 Oracle Database JDBC Developer's Guide and Reference

Update Batching

3.

If you send an execute-update call to the database at this point, then no data will be sent to the database, and the call will return 0.
// No data is sent to the database by this call to executeUpdate System.out.println ("Number of rows updated so far: " + ps.executeUpdate ());

4.

If you enter a set of input values for a second row and an execute-update, then the number of batch calls to executeUpdate will be equal to the batch value of 2. The data will be sent to the database, and both rows will be inserted in a single round-trip.
ps.setInt (1, 11); ps.setString (2, "Applications"); ps.setString (3, "Indonesia"); int rows = ps.executeUpdate (); System.out.println ("Number of rows updated now: " + rows); ps.close ();

Checking the Batch Value

To check the overall connection batch value of an Oracle connection instance, use the OracleConnection class getDefaultExecuteBatch method:
Integer batch_val = ((OracleConnection)conn).getDefaultExecuteBatch();

To check the particular statement batch value of an Oracle prepared statement, use the OraclePreparedStatement class getExecuteBatch method:
Integer batch_val = ((OraclePreparedStatement)ps).getExecuteBatch();

Note:

If no statement batch value has been set, then getExecuteBatch will return the connection batch value.

Overriding the Batch Value

If you want to process accumulated operations before the batch value in effect is reached, then use the sendBatch method of the OraclePreparedStatement object. For this example, presume you set the connection batch value to 20. This sets the default batch value for all prepared statement objects associated with the connection to 20. You can accomplish this by casting your connection to OracleConnection and applying the setDefaultExecuteBatch method for the connection, as follows:
((OracleConnection)conn).setDefaultExecuteBatch (20);

Override the batch value as follows:

1.

Write your prepared statement, specify input values for the first row, and then process the statement, as follows:
PreparedStatement ps = conn.prepareStatement ("insert into dept values (?, ?, ?)"); ps.setInt (1, 32); ps.setString (2, "Oracle");

Performance Extensions 23-5

Update Batching

ps.setString (3, "USA"); System.out.println (ps.executeUpdate ());

The batch is not processed at this point. The ps.executeUpdate method returns 0.
2.

If you enter a set of input values for a second operation and call executeUpdate again, then the data will still not be sent to the database, because the batch value in effect for the statement is the connection batch value, which is 20.
ps.setInt (1, 33); ps.setString (2, "Applications"); ps.setString (3, "Indonesia"); // this batch is still not executed at this point int rows = ps.executeUpdate (); System.out.println ("Number of rows updated before calling sendBatch: " + rows);

Note that the value of rows in the println statement is 0.

3.

If you apply the sendBatch method at this point, then the two previously batched operations will be sent to the database in a single round-trip. The sendBatch method also returns the total number of updated rows. This property of sendBatch is used by println to print the number of updated rows.
// Execution of both previously batched executes will happen // at this point. The number of rows updated will be // returned by sendBatch. rows = ((OraclePreparedStatement)ps).sendBatch (); System.out.println ("Number of rows updated by calling sendBatch: " + rows); ps.close ();

Committing the Changes in Oracle Batching

After you process the batch, you must still commit the changes, presuming auto-commit is disabled as recommended. Calling commit on the connection object in Oracle batching not only commits operations in batches that have been processed, but also issues an implicit sendBatch call to process all pending batches. So commit effectively commits changes for all operations that have been added to a batch.

Update Counts in Oracle Batching

In a non-batching situation, the executeUpdate method of an OraclePreparedStatement object will return the number of database rows affected by the operation. In an Oracle batching situation, this method returns the number of rows affected at the time the method is invoked, as follows:

If an executeUpdate call results in the operation being added to the batch, then the method returns a value of 0, because nothing was written to the database yet.

23-6 Oracle Database JDBC Developer's Guide and Reference

Update Batching

If an executeUpdate call results in the batch value being reached and the batch being processed, then the method will return the total number of rows affected by all operations in the batch.

Similarly, the sendBatch method of an OraclePreparedStatement object returns the total number of rows affected by all operations in the batch. Example 231 illustrates the use of Oracle update batching.
Example 231 Oracle Update Batching

The following example illustrates how you use the Oracle update batching feature. It assumes you have imported the oracle.driver.* interfaces.
... OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:oracle:oci); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection(); conn.setAutoCommit(false); PreparedStatement ps = conn.prepareStatement("insert into dept values (?, ?, ?)"); //Change batch size for this statement to 3 ((OraclePreparedStatement)ps).setExecuteBatch (3); ps.setInt(1, 23); ps.setString(2, "Sales"); ps.setString(3, "USA"); ps.executeUpdate(); //JDBC queues this for later execution ps.setInt(1, 24); ps.setString(2, "Blue Sky"); ps.setString(3, "Montana"); ps.executeUpdate(); //JDBC queues this for later execution ps.setInt(1, 25); ps.setString(2, "Applications"); ps.setString(3, "India"); ps.executeUpdate(); //The queue size equals the batch value of 3 //JDBC sends the requests to the database ps.setInt(1, 26); ps.setString(2, "HR"); ps.setString(3, "Mongolia"); ps.executeUpdate(); //JDBC queues this for later execution ((OraclePreparedStatement)ps).sendBatch(); // JDBC sends the queued request conn.commit(); ps.close(); ...

Performance Extensions 23-7

Update Batching

Note: Updates deferred through batching can affect the results of other queries. In the following example, if the first query is deferred due to batching, then the second will return unexpected results:
UPDATE emp SET name = "Sue" WHERE name = "Bob"; SELECT name FROM emp WHERE name = "Sue";

Error Reporting in Oracle Update Batching

If any one of the batched operations fails to complete successfully or attempts to return a result set during an executeBatch call, then the processing stops and a java.sql.BatchUpdateException is generated. If the exception is raised, you can call the getUpdateCounts method on the BatchUpdateException object to retrieve the update count. This method returns an int array of update counts, just as the executeBatch method does. In Oracle Database 10g release 2 (10.2), the integer array returned contains n Statement.EXECUTE_FAILED entries, where n is the size of the batch. However, this does not indicate where in the batch the error occurred. The only option you have is to rollback the transaction. In Oracle Database 11g Release 1 (11.1), the integer array returned contains n Statement.SUCCESS_NO_INFO entries, where n is the number of elements in the batch that have been successfully executed.

Note:

The execution of the batch always stops with the first element of the batch that generates an error.

Standard Update Batching

Oracle implements the standard update batching model according to the JDBC 2.0 specification. This model, unlike the Oracle update batching model, depends on explicitly adding statements to the batch using an addBatch method and explicitly processing the batch using an executeBatch method. In the Oracle model, you call executeUpdate as in a non-batching situation, but whether an operation is added to the batch or the whole batch is processed is typically determined implicitly, depending on whether a pre-determined batch value is reached.
Note:

Do not mix standard update batching with Oracle update batching in the same application. Oracle JDBC driver will throw exceptions when these are mixed. Disable auto-commit mode if you use either update batching model. In case an error occurs while you are processing a batch, this provides you the option of committing or rolling back the operations that ran successfully prior to the error.

Limitations in the Oracle Implementation of Standard Batching

This section discusses the limitations and implementation details regarding the Oracle implementation of standard update batching.

23-8 Oracle Database JDBC Developer's Guide and Reference

Update Batching

In Oracle JDBC applications, update batching is intended for use with prepared statements that are being processed repeatedly with different sets of bind values. The Oracle implementation of standard update batching does not implement true batching for generic statements and callable statements. Even though Oracle JDBC supports the use of standard batching for Statement and CallableStatement objects, you are unlikely to see performance improvement.

Adding Operations to the Batch

When any statement object is first created, its statement batch is empty. Use the standard addBatch method to add an operation to the statement batch. This method is specified in the standard java.sql.Statement, PreparedStatement, and CallableStatement interfaces, which are implemented by the oracle.jdbc.OracleStatement, OraclePreparedStatement, and OracleCallableStatement interfaces, respectively. For a Statement object, the addBatch method takes a Java String with a SQL operation as input. For example:
... Statement stmt = conn.createStatement(); stmt.addBatch("INSERT INTO emp VALUES(1000, 'Joe Jones')"); stmt.addBatch("INSERT INTO dept VALUES(260, 'Sales')"); stmt.addBatch("INSERT INTO emp_dept VALUES(1000, 260)"); ...

At this point, three operations are in the batch.

Note:

Remember, however, that in the Oracle implementation of standard update batching, you will probably see no performance improvement in batching generic statements.

For prepared statements, update batching is used to batch multiple runs of the same statement with different sets of bind parameters. For a PreparedStatement or OraclePreparedStatement object, the addBatch method takes no input. It simply adds the operation to the batch using the bind parameters last set by the appropriate setXXX methods. This is also true for CallableStatement or OracleCallableStatement objects, but remember that in the Oracle implementation of standard update batching, you will probably see no performance improvement in batching callable statements. For example:
... PreparedStatement pstmt = conn.prepareStatement("INSERT INTO employees VALUES(?, ?)"); pstmt.setInt(1, 2000); pstmt.setString(2, "Milo Mumford"); pstmt.addBatch(); pstmt.setInt(1, 3000); pstmt.setString(2, "Sulu Simpson"); pstmt.addBatch(); ...

At this point, two operations are in the batch.

Performance Extensions 23-9

Update Batching

Because a batch is associated with a single prepared statement object, you can batch only repeated runs of a single prepared statement, as in this example.

Processing the Batch

To process the current batch of operations, use the executeBatch method of the statement object. This method is specified in the standard Statement interface, which is extended by the standard PreparedStatement and CallableStatement interfaces. Following is an example that repeats the prepared statement addBatch calls shown previously and then processes the batch:
... PreparedStatement pstmt = conn.prepareStatement("INSERT INTO employees VALUES(?, ?)"); pstmt.setInt(1, 2000); pstmt.setString(2, "Milo Mumford"); pstmt.addBatch(); pstmt.setInt(1, 3000); pstmt.setString(2, "Sulu Simpson"); pstmt.addBatch(); int[] updateCounts = pstmt.executeBatch(); ...

Starting from Oracle Database 11g Release 1 (11.1), the executeBatch method has been improved so that when an error occurs in the middle of the batch execution, the BatchUpdateExecution exception that is thrown contains the position of the error in the batch. The BatchUpdateExecution.getUpdateCounts method returns an array of int containing the update counts for the updates that were executed successfully before this error occurred. So if an error occurs in the 5th element of the batch, then the size of the array returned is 4 and each value is Statement.SUCCESS_NO_INFO.

Committing the Changes in the Oracle Implementation of Standard Batching

After you process the batch, you must still commit the changes, presuming auto-commit is disabled as recommended. Calling commit, commits non-batched operations and batched operations for statement batches that have been processed, but for the Oracle implementation of standard batching, has no effect on pending statement batches that have not been processed.

Clearing the Batch

To clear the current batch of operations instead of processing it, use the clearBatch method of the statement object. This method is specified in the standard Statement interface, which is extended by the standard PreparedStatement and CallableStatement interfaces. Keep the following things in mind:

When a batch is processed, operations are performed in the order in which they were batched. After calling addBatch, you must call either executeBatch or clearBatch before a call to executeUpdate, otherwise there will be a SQL exception.

23-10 Oracle Database JDBC Developer's Guide and Reference

Update Batching

A clearBatch or executeBatch call resets the statement batch to empty. The statement batch is not reset to empty if the connection receives a ROLLBACK request. You must explicitly call clearBatch to reset it.
Note:

If you are using Oracle update batching in an Oracle Database 11g, then you do not have to clear your batches explicitly in the code after a rollback. However, it is OK to invoke clearBatch method after a rollback. If you are using Oracle update batching in an earlier release, then you have to invoke clearBatch method to clear your batches explicitly after a rollback.

Invoking clearBatch method after a rollback works for all releases. An executeBatch call closes the current result set of the statement object, if one exists. Nothing is returned by the clearBatch method.

Following is an example that repeats the prepared statement addBatch calls shown previously but then clears the batch under certain circumstances:
... PreparedStatement pstmt = conn.prepareStatement("INSERT INTO employees VALUES(?, ?)"); pstmt.setInt(1, 2000); pstmt.setString(2, "Milo Mumford"); pstmt.addBatch(); pstmt.setInt(1, 3000); pstmt.setString(2, "Sulu Simpson"); pstmt.addBatch(); if (...condition...) { int[] updateCounts = pstmt.executeBatch(); ... } else { pstmt.clearBatch(); ... }

Update Counts in the Oracle Implementation of Standard Batching

If a statement batch is processed successfully, then the integer array, or update counts array, returned by the statement executeBatch call will always have one element for each operation in the batch. In the Oracle implementation of standard update batching, the values of the array elements are as follows:

For a prepared statement batch, it is not possible to know the number of rows affected in the database by each individual statement in the batch. Therefore, all array elements have a value of -2. According to the JDBC 2.0 specification, a value

Performance Extensions

23-11

Update Batching

of -2 indicates that the operation was successful but the number of rows affected is unknown.

For a generic statement batch the array contains the actual update counts indicating the number of rows affected by each operation. The actual update counts can be provided only in the case of generic statements in Oracle implementation of standard batching. For a callable statement batch, the server always returns the value 1 as the update count, irrespective of the number rows affected by each operation.

In your code, upon successful processing of a batch, you should be prepared to handle either -2, 1, or true update counts in the array elements. For a successful batch processing, the array contains either all -2, 1, or all positive integers. Example 232 illustrates the use of standard update batching.
Example 232 Standard Update Batching

This example combines the sample fragments in the previous sections, accomplishing the following steps:
1. 2. 3. 4. 5.

Disabling auto-commit mode, which you should always do when using either update batching model Creating a prepared statement object Adding operations to the batch associated with the prepared statement object Processing the batch Committing the operations from the batch

conn.setAutoCommit(false); PreparedStatement pstmt = conn.prepareStatement("INSERT INTO employees VALUES(?, ?)"); pstmt.setInt(1, 2000); pstmt.setString(2, "Milo Mumford"); pstmt.addBatch(); pstmt.setInt(1, 3000); pstmt.setString(2, "Sulu Simpson"); pstmt.addBatch(); int[] updateCounts = pstmt.executeBatch(); conn.commit(); pstmt.close(); ...

You can process the update counts array to determine if the batch processed successfully.

Error Handling in the Oracle Implementation of Standard Batching

If any one of the batched operations fails to complete successfully or attempts to return a result set during an executeBatch call, then the processing stops and a java.sql.BatchUpdateException is generated.

23-12 Oracle Database JDBC Developer's Guide and Reference

Update Batching

After a batch exception, the update counts array can be retrieved using the getUpdateCounts method of the BatchUpdateException object. This returns an int array of update counts, just as the executeBatch method does. In the Oracle implementation of standard update batching, contents of the update counts array are as follows, after a batch is processed:

For a prepared statement batch, it is not possible to know which operation failed. The array has one element for each operation in the batch, and each element has a value of -3. According to the JDBC 2.0 specification, a value of -3 indicates that an operation did not complete successfully. In this case, it was presumably just one operation that actually failed, but because the JDBC driver does not know which operation that was, it labels all the batched operations as failures. You should always perform a ROLLBACK operation in this situation.

For a generic statement batch or callable statement batch, the update counts array is only a partial array containing the actual update counts up to the point of the error. The actual update counts can be provided because Oracle JDBC cannot use true batching for generic and callable statements in the Oracle implementation of standard update batching. For example, if there were 20 operations in the batch, the first 13 succeeded, and the 14th generated an exception, then the update counts array will have 13 elements, containing actual update counts of the successful operations. You can either commit or roll back the successful operations in this situation, as you prefer.

In your code, upon failed processing of a batch, you should be prepared to handle either -3 or true update counts in the array elements when an exception occurs. For a failed batch processing, you will have either a full array of -3 or a partial array of positive integers.

Intermixing Batched Statements and Non-Batched Statements

You cannot call executeUpdate for regular, non-batched processing of an operation if the statement object has a pending batch of operations. However, you can intermix batched operations and non-batched operations in a single statement object if you process non-batched operations either prior to adding any operations to the statement batch or after processing the batch. Essentially, you can call executeUpdate for a statement object only when its update batch is empty. If the batch is non-empty, then an exception will be generated. For example, it is legal to have a sequence, such as the following:
... PreparedStatement pstmt = conn.prepareStatement("INSERT INTO employees VALUES(?, ?)"); pstmt.setInt(1, 2000); pstmt.setString(2, "Milo Mumford"); int scount = pstmt.executeUpdate(); pstmt.setInt(1, 3000); pstmt.setString(2, "Sulu Simpson"); pstmt.addBatch(); pstmt.setInt(1, 4000); pstmt.setString(2, "Stan Leland"); // OK; no operations in pstmt batch

// Now start a batch

Performance Extensions

23-13

Update Batching

pstmt.addBatch(); int[] bcounts = pstmt.executeBatch(); pstmt.setInt(1, 5000); pstmt.setString(2, "Amy Feiner"); int scount = pstmt.executeUpdate(); ... // OK; pstmt batch was executed

Intermixing non-batched operations on one statement object and batched operations on another statement object within your code is permissible. Different statement objects are independent of each other with regards to update batching operations. A COMMIT request will affect all non-batched operations and all successful operations in processed batches, but will not affect any pending batches.

Premature Batch Flush

Premature batch flush happens due to a change in cached metadata. Cached metadata can be changed due to various reasons, such as the following:

The initial bind was null and the following bind is not null A scalar type is initially bound as string and then bound as scalar type or the reverse

The premature batch flush count is summed to the return value of the next executeUpdate or sendBatch method. The old functionality lost all these batch flush values which can be obtained now. To switch back to the old functionality, you can set the AccumulateBatchResult property to false, as follows:
java.util.Properties info = new java.util.Properties(); info.setProperty("user", "SCOTT"); info.setProperty("passwd", "TIGER"); // other properties ... // property: batch flush type info.setProperty("AccumulateBatchResult", "false"); OracleDataSource ods = new OracleDataSource(); ods.setConnectionProperties(info); ods.setURL("jdbc:oracle:oci:@""); Connection conn = ods.getConnection();

Note: The AccumulateBatchResult property is set to true by default.

Example 233 illustrates premature batch flushing.

Example 233 Premature Batch Flushing

((OraclePreparedStatement)pstmt).setExecuteBatch (2); pstmt.setNull (1, OracleTypes.NUMBER); pstmt.setString (2, "test11");

23-14 Oracle Database JDBC Developer's Guide and Reference

Additional Oracle Performance Extensions

int count = pstmt.executeUpdate (); // returns 0 /* * Premature batch flush happens here. */ pstmt.setInt (1, 22); pstmt.setString (2, "test22"); int count = pstmt.executeUpdate (); // returns 0 pstmt.setInt (1, 33); pstmt.setString (2, "test33"); /* * returns 3 with the new batching scheme where as, * returns 2 with the old batching scheme. */ int count = pstmt.executeUpdate ();

Additional Oracle Performance Extensions

In addition to update batching, Oracle JDBC drivers support the following extensions that improve performance by reducing round-trips to the database:

Prefetching rows This reduces round-trips to the database by fetching multiple rows of data each time data is fetched. The extra data is stored in client-side buffers for later access by the client. The number of rows to prefetch can be set as desired.

Specifying column types This avoids an inefficiency in the standard JDBC protocol for performing and returning the results of queries.

Suppressing database metadata TABLE_REMARKS columns This avoids an expensive outer join operation.

Oracle provides several extensions to connection properties objects to support these performance extensions. These extensions enable you to set the remarksReporting flag and default values for row prefetching and update batching. This section covers the following topics:

Oracle Row-Prefetching Limitations Defining Column Types DatabaseMetaData TABLE_REMARKS Reporting

Oracle Row-Prefetching Limitations

There is no maximum prefetch setting. The default value of 10. Larger or smaller values may be appropriate depending on the number of rows and columns expected from the query. You can set the default connection row-prefetch value using a Properties object. When a statement object is created, it receives the default row-prefetch setting from the associated connection. Subsequent changes to the default connection row-prefetch setting will have no effect on the statement row-prefetch setting. If a column of a result set is of data type LONG, LONG RAW or LOBs returned through the data interface, that is, the streaming types, then JDBC changes the statement
Performance Extensions 23-15

Additional Oracle Performance Extensions

row-prefetch setting to 1, even if you never actually read a value of either of these types. Setting the prefetch size can effect the performance of an application. Increasing the prefetch size will reduce the number of round-trips required to get all the data, but will increase memory usage. This will depend on the number and size of the columns in the query and the number of rows expected to be returned. It will also depend on the memory and CPU loading of the JDBC client machine. The optimum for a standalone client application will be different from a heavily loaded application server. Please consider also the speed and latency of the network connection.
Note:

Starting from Oracle Database 11g Release 1 (11.1), the thin driver can fetch the first prefetch_size number of rows from the server in the very first roundtrip. This saves one round-trip in select statements.

If you are migrating an application from earlier releases of Oracle JDBC drivers to 10g release 1 (10.1) or later releases of Oracle JDBC drivers, then you should revisit the optimizations that you had done earlier, because the memory usage and performance characteristics may have changed substantially. A common situation that you may encounter is, say, you have a query that selects a unique key. The query will return only zero or one row. Setting the prefetch size to 1 will decrease memory and CPU usage and cannot increase round-trips. However, you must be careful to avoid the error of requesting an extra fetch by writing while(rs.next()) instead of if(rs.next()). If you are using the JDBC Thin driver, then in the case where only zero or one row is expected, use the useFetchSizeWithLongColumn connection property, because it will perform PARSE, EXECUTE, and FETCH in a single round-trip. Tuning of the prefetch size should be done along with tuning of memory management in your JVM under realistic loads of the actual application.
Note:

Do not mix the JDBC 2.0 fetch size application programming interface (API) and the Oracle row-prefetching API in your application. You can use one or the other, but not both. Be aware that setting the Oracle fetch size value can affect not only queries, but also explicitly refetching rows in a result set through the result set refreshRow method, which is relevant for scroll-sensitive/read-only, scroll-sensitive/updatable, and scroll-insensitive/updatable result sets, and the window size of a scroll-sensitive result set, affecting how often automatic refetches are performed. However, the Oracle fetch size value will be overridden by any setting of the fetch size. "Supported Connection Properties" on page 8-8

See Also:

Defining Column Types

The implementation of defineColumnType changed significantly in Oracle Database 10g. Previously, defineColumnType was used both as a performance optimization and to force data type conversion. In previous releases, all of the drivers benefited
23-16 Oracle Database JDBC Developer's Guide and Reference

Additional Oracle Performance Extensions

from calls to defineColumnType. In Oracle Database 10g, the JDBC Thin driver no longer needs the information provided. The JDBC Thin driver achieves maximum performance without calls to defineColumnType. The JDBC Oracle Call Interface (OCI) and server-side internal drivers still get better performance when the application uses defineColumnType. If your code is used with both the JDBC Thin and OCI drivers, you can disable the defineColumnType method when using the Thin by setting the connection property disableDefineColumnType to true. Doing this makes defineColumnType have no effect. Do not set this connection property to true when using the JDBC OCI or server-side internal drivers. You can also use defineColumnType to control how much memory the client-side allocates or to limit the size of variable-length data. Follow these general steps to define column types for a query:
1. 2. 3.

If necessary, cast your statement object to OracleStatement, OraclePreparedStatement, or OracleCallableStatement, as applicable. If necessary, use the clearDefines method of your Statement object to clear any previous column definitions for this Statement object. On each column, call the defineColumnType method of your Statement object, passing it these parameters:

Column index (integer) Type code (integer) Use the static constants of the java.sql.Types class or oracle.jdbc.OracleTypes class, such as Types.INTEGER, Types.FLOAT, Types.VARCHAR, OracleTypes.VARCHAR, and OracleTypes.ROWID. Type codes for standard types are identical in these two classes.

Type name (string) For structured objects, object references, and arrays, you must also specify the type name. For example, Employee, EmployeeRef, or EmployeeArray.

Maximum field size (integer) Optionally specify a maximum data length for this column. You cannot specify a maximum field size parameter if you are defining the column type for a structured object, object reference, or array. If you try to include this parameter, it will be ignored.

Form-of-use (short) Optionally specify a form of use for the column. This can be OraclePreparedStatement.FORM_CHAR to use the database character set or OraclePreparedStatement.FORM_NCHAR to use the national character set. If this parameter is omitted, the default is FORM_CHAR.

For example, assuming stmt is an Oracle statement, use:

stmt.defineColumnType(column_index, typeCode);

or, if the column is VARCHAR or equivalent and you know the length limit:
stmt.defineColumnType(column_index, typeCode, max_size);

Performance Extensions

23-17

Additional Oracle Performance Extensions

or, for an NVARCHAR column where the original maximum length is desired and conversion to the database character set is requested:
stmt.defineColumnType(column_index, typeCode, 0, OraclePreparedStatement.FORM_CHAR );

or, for structured object, object reference, and array columns:

stmt.defineColumnType(column_index, typeCode, typeName);

Set a maximum field size if you do not want to receive the full default length of the data. Calling the setMaxFieldSize method of the standard JDBC Statement class sets a restriction on the amount of data returned. Specifically, the size of the data returned will be the minimum of the following:

The maximum field size set in defineColumnType The maximum field size set in setMaxFieldSize The natural maximum size of the data type

After you complete these steps, use the executeQuery method of the statement to perform the query.
Note:

It is no longer necessary to specify a data type for each column of the expected result set.

Example 234 illustrates the use of this feature. It assumes you have imported the oracle.jdbc.* interfaces.
Example 234 Defining Column Types

OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:oracle:thin:@localhost:1502:orcl"); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection(); Statement stmt = conn.createStatement(); // Allocate only 2 chars for this column (truncation will happen) ((OracleStatement)stmt).defineColumnType(1, Types.VARCHAR, 2); ResultSet rset = stmt.executeQuery("select ename from emp"); while (rset.next() ) System.out.println(rset.getString(1)); stmt.close();

As this example shows, you must cast the Statement object, stmt, to OracleStatement in the invocation of the defineColumnType method. The createStatement method of the connection returns an object of type java.sql.Statement, which does not have the defineColumnType and clearDefines methods. These methods are provided only in the OracleStatement implementation. The define-extensions use JDBC types to specify the desired types. The allowed define types for columns depend on the internal Oracle type of the column. All columns can be defined to their natural JDBC types. In most cases, they can be defined to the Types.CHAR or Types.VARCHAR type code.

23-18 Oracle Database JDBC Developer's Guide and Reference

Additional Oracle Performance Extensions

Table 231 lists the valid column definition arguments you can use in the defineColumnType method.
Table 231 Valid Column Type Specifications You can use defineColumnType to define it as: BIGINT, TINYINT, SMALLINT, INTEGER, FLOAT, REAL, DOUBLE, NUMERIC, DECIMAL, CHAR, VARCHAR CHAR, VARCHAR CHAR, VARCHAR, LONGVARCHAR LONGVARBINARY, VARBINARY, BINARY VARBINARY, BINARY DATE, TIME, TIMESTAMP, CHAR, VARCHAR ROWID VARBINARY, BINARY LONG, CHAR, VARCHAR

If the column has Oracle SQL type: NUMBER, VARNUM CHAR, VARCHAR2 LONG LONGRAW RAW DATE ROWID BLOB CLOB

It is always valid to use defineColumnType with the original data type of the column.

DatabaseMetaData TABLE_REMARKS Reporting

The getColumns, getProcedureColumns, getProcedures, and getTables methods of the database metadata classes are slow if they must report TABLE_REMARKS columns, because this necessitates an expensive outer join. For this reason, the JDBC driver does not report TABLE_REMARKS columns by default. You can enable TABLE_REMARKS reporting by passing a true argument to the setRemarksReporting method of an OracleConnection object. Equivalently, instead of calling setRemarksReporting, you can set the remarksReporting Java property if you use a Java Properties object in establishing the connection. If you are using a standard java.sql.Connection object, you must cast it to OracleConnection to use setRemarksReporting. Example 235 illustrates how to enable TABLE_REMARKS reporting.
Example 235 TABLE_REMARKS Reporting

Assuming conn is the name of your standard Connection object, the following statement enables TABLE_REMARKS reporting.
( (oracle.jdbc.OracleConnection)conn ).setRemarksReporting(true);

Considerations for getColumns By default, the getColumns method does not retrieve information about the columns if a synonym is specified. To enable the retrieval of information if a synonym is specified, you must call the setIncludeSynonyms method on the connection as follows:
( (oracle.jdbc.driver.OracleConnection)conn ).setIncludeSynonyms(true)

Performance Extensions

23-19

Additional Oracle Performance Extensions

This will cause all subsequent getColumns method call on the connection to include synonyms. This is similar to setRemarksReporting. Alternatively, you can set the includeSynonyms connection property. This is similar to the remarksReporting connection property. However, you need to bear in mind that if includeSynonyms is true, then the name of the object returned in the table_name column will be the synonym name, if a synonym exists. This is true even if you pass the table name to getColumns. Considerations for getProcedures and getProcedureColumns Methods According to JDBC versions 1.1 and 1.2, the methods getProcedures and getProcedureColumns treat the catalog, schemaPattern, columnNamePattern, and procedureNamePattern parameters in the same way. In the Oracle definition of these methods, the parameters are treated differently:

catalog Oracle does not have multiple catalogs, but it does have packages. Consequently, the catalog parameter is treated as the package name. This applies both on input, which is the catalog parameter, and the output, which is the catalog column in the returned ResultSet. On input, the construct " ", which is an empty string, retrieves procedures and arguments without a package, that is, standalone objects. A null value means to drop from the selection criteria, that is, return information about both standalone and packaged objects. That is, it has the same effect as passing in %. Otherwise the catalog parameter should be a package name pattern, with SQL wild cards, if desired.

schemaPattern All objects within Oracle must have a schema, so it does not make sense to return information for those objects without one. Thus, the construct " ", which is an empty string, is interpreted on input to mean the objects in the current schema, that is, the one to which you are currently connected. To be consistent with the behavior of the catalog parameter, null is interpreted to drop the schema from the selection criteria. That is, it has the same effect as passing in %. It can also be used as a pattern with SQL wild cards.

procedureNamePattern and columnNamePattern The empty string (" ") does not make sense for either parameter, because all procedures and arguments must have names. Thus, the construct " " will raise an exception. To be consistent with the behavior of other parameters, null has the same effect as passing in %.

23-20 Oracle Database JDBC Developer's Guide and Reference

24
OCI Connection Pooling
The Java Database Connectivity (JDBC) Oracle Call Interface (OCI) driver connection pooling functionality is part of the JDBC client. This functionality is provided by the OracleOCIConnectionPool class. A JDBC application can have multiple pools at the same time. Multiple pools can correspond to multiple application servers or pools to different data sources. The connection pooling provided by the JDBC OCI driver enables applications to have multiple logical connections, all using a small set of physical connections. Each call on a logical connection gets routed on to the physical connection that is available at the time of call. This chapter contains the following sections:

OCI Driver Connection Pooling: Background OCI Driver Connection Pooling and Shared Servers Compared Defining an OCI Connection Pool Connecting to an OCI Connection Pool Sample Code for OCI Connection Pooling Statement Handling and Caching JNDI and the OCI Connection Pool
Note:

Use OCI connection pooling if you need session multiplexing. Otherwise, Oracle recommends using the Implicit Connection Cache functionality.

OCI Driver Connection Pooling: Background

The Oracle JDBC OCI driver provides several transaction monitor capabilities, such as the fine-grained management of Oracle sessions and connections. It is possible for a high-end application server or transaction monitor to multiplex several sessions over fewer physical connections on a call-level basis, thereby achieving a high degree of scalability by pooling of connections and back-end Oracle server processes. The connection pooling provided by the OracleOCIConnectionPool interface simplifies the session/connection separation interface hiding the management of the physical connection pool. The Oracle sessions are the OracleOCIConnection objects obtained from OracleOCIConnectionPool. The connection pool itself is usually configured with a much smaller shared pool of physical connections, translating to a back-end server pool containing an identical number of dedicated server processes.

OCI Connection Pooling 24-1

OCI Driver Connection Pooling and Shared Servers Compared

Note that many more Oracle sessions can be multiplexed over this pool of fewer shared connections and back-end Oracle processes.

OCI Driver Connection Pooling and Shared Servers Compared

In some ways, what OCI driver connection pooling offers on the middle tier is similar to what shared server processes offer on the back-end. OCI driver connection pooling makes a dedicated server instance behaves as an shared instance by managing the session multiplexing logic on the middle tier. Therefore, the pooling of dedicated server processes and incoming connections into the dedicated server processes is controlled by the OCI connection pool on the middle tier. The main difference between OCI connection pooling and shared servers is that in case of shared servers, the connection from the client is typically to a dispatcher in the database instance. The dispatcher is responsible for directing the client request to an appropriate shared server. On the other hand, the physical connection from the OCI connection pool is established directly from the middle tier to the Oracle dedicated server process in the back-end server pool. Note that OCI connection pool is mainly beneficial only if the middle tier is multithreaded. Each thread could maintain a session to the database. The actual connections to the database are maintained by OracleOCIConnectionPool, and these connections, including the pool of dedicated database server processes, are shared among all the threads in the middle tier.

Defining an OCI Connection Pool

An OCI connection pool is created at the beginning of the application. Creating connections from a pool is quite similar to creating connections using the OracleDataSource class. The oracle.jdbc.pool.OracleOCIConnectionPool class, which extends the OracleDataSource class, is used to create OCI connection pools. From an OracleOCIConnectionPool instance, you can obtain logical connection objects. These connection objects are of the OracleOCIConnection class type. This class implements the OracleConnection interface. The Statement objects you create from the OracleOCIConnection instance have the same fields and methods as OracleStatement objects you create from OracleConnection instances. The following code shows header information for the OracleOCIConnectionPool class:
/* * @param us ConnectionPool user-id. * @param p ConnectionPool password * @param name logical name of the pool. This needs to * tnsnames.ora configuration file. @param config (optional) Properties of the pool, if suffice. Default connection configuration incr=0 Please refer setPoolConfig for property

be one in the the default does not is min =1, max=1, names.

Since this is optional, pass null if the default configuration suffices. * @return * * Notes: Choose a userid and password that can act as proxy for the users

24-2 Oracle Database JDBC Developer's Guide and Reference

Defining an OCI Connection Pool

in the getProxyConnection() method. If config is null, effect CONNPOOL_MIN_LIMIT CONNPOOL_MAX_LIMIT CONNPOOL_INCREMENT then the following default values will take = 1 = 1 = 0

*/ public synchronized OracleOCIConnectionPool (String user, String password, throws SQLException

String name, Properties config)

/* * This will use the user-id, password and connection pool name values set LATER using the methods setUser, setPassword, setConnectionPoolName. * @return * * Notes: No OracleOCIConnection objects can be created on this class unless the methods setUser, setPassword, setPoolConfig are invoked. When invoking the setUser, setPassword later, choose a userid and password that can act as proxy for the users in the getProxyConnection() method.

* */ public synchronized OracleOCIConnectionPool () throws SQLException

Importing the oracle.jdbc.pool and oracle.jdbc.oci Packages Before you create an OCI connection pool, import the following to have Oracle OCI connection pooling functionality:
import oracle.jdbc.pool.*; import oracle.jdbc.oci.*;

Creating an OCI Connection Pool The following code show how you create an instance of the OracleOCIConnectionPool class called cpool:
OracleOCIConnectionPool cpool = new OracleOCIConnectionPool ("SCOTT", "TIGER", "jdbc:oracle:oci:@(description=(address=(host= myhost)(protocol=tcp)(port=1521))(connect_data=(INSTANCE_NAME=orcl)))", poolConfig);

poolConfig is a set of properties which specify the connection pool. If poolConfig is null, then the default values are used. For example, consider the following:

poolConfig.put (OracleOCIConnectionPool.CONNPOOL_MIN_LIMIT, "4"); poolConfig.put (OracleOCIConnectionPool.CONNPOOL_MAX_LIMIT, "10");

OCI Connection Pooling 24-3

Defining an OCI Connection Pool

poolConfig.put (OracleOCIConnectionPool.CONNPOOL_INCREMENT, "2");

As an alternative to the constructor call, you can create an instance of the OracleOCIConnectionPool class using individual methods to specify the user, password, and connection string.
OracleOCIConnectionPool cpool = new OracleOCIConnectionPool ( ); cpool.setUser("SCOTT"); cpool.setPassword("TIGER"); cpool.setURL("jdbc:oracle:oci:@(description=(address=(host= myhost)(protocol=tcp)(port=1521))(connect_data=(INSTANCE_NAME=orcl)))"); cpool.setPoolConfig(poolConfig); // In case you want to specify a different // configuration other than the default // values.

Setting the OCI Connection Pool Parameters The connection pool configuration is determined by the following OracleOCIConnectionPool class attributes:

CONNPOOL_MIN_LIMIT Specifies the minimum number of physical connections that can be maintained by the pool.

CONNPOOL_MAX_LIMIT Specifies the maximum number of physical connections that can be maintained by the pool.

CONNPOOL_INCREMENT Specifies the incremental number of physical connections to be opened when all the existing ones are busy and a call needs one more connection; the increment is done only when the total number of open physical connections is less than the maximum number that can be opened in that pool.

CONNPOOL_TIMEOUT Specifies how much time must pass before an idle physical connection is disconnected; this does not affect a logical connection.

CONNPOOL_NOWAIT Specifies, if enabled, that an error is returned if a call needs a physical connection while the maximum number of connections in the pool are busy. If disabled, a call waits until a connection is available. Once this attribute is set to true, it cannot be reset to false.

You can configure all of these attributes dynamically. Therefore, an application has the flexibility of reading the current load, that is number of open connections and number of busy connections, and adjusting these attributes appropriately, using the setPoolConfig method.
Note: The default values for the CONNPOOL_MIN_LIMIT, CONNPOOL_MAX_LIMIT, and CONNPOOL_INCREMENT parameters are 1, 1, and 0, respectively.

24-4 Oracle Database JDBC Developer's Guide and Reference

Defining an OCI Connection Pool

The setPoolConfig method is used to configure OCI connection pool properties. The following is a typical example of how the OracleOCIConnectionPool class attributes can be set:
... java.util.Properties p = new java.util.Properties( ); p.put (OracleOCIConnectionPool.CONNPOOL_MIN_LIMIT, "1"); p.put (OracleOCIConnectionPool.CONNPOOL_MAX_LIMIT, "5"); p.put (OracleOCIConnectionPool.CONNPOOL_INCREMENT, "2"); p.put (OracleOCIConnectionPool.CONNPOOL_TIMEOUT, "10"); p.put (OracleOCIConnectionPool.CONNPOOL_NOWAIT, "true"); cpool.setPoolConfig(p); ...

Observe the following rules when setting these attributes:

CONNPOOL_MIN_LIMIT, CONNPOOL_MAX_LIMIT, and CONNPOOL_INCREMENT are mandatory CONNPOOL_MIN_LIMIT must be a value greater than zero CONNPOOL_MAX_LIMIT must be a value greater than or equal to CONNPOOL_MIN_LIMIT plus CONNPOOL_INCREMENT CONNPOOL_INCREMENT must be a value greater than or equal to zero CONNPOOL_TIMEOUT must be a value greater than zero CONNPOOL_NOWAIT must be true or false
See Also:

Oracle Call Interface Programmer's Guide

Checking the OCI Connection Pool Status To check the status of the connection pool, use the following methods from the OracleOCIConnectionPool class:

int getMinLimit() Retrieves the minimum number of physical connections that can be maintained by the pool.

int getMaxLimit() Retrieves the maximum number of physical connections that can be maintained by the pool.

int getConnectionIncrement() Retrieves the incremental number of physical connections to be opened when all the existing ones are busy and a call needs a connection.

int getTimeout() Retrieves the specified time (in seconds) that a physical connection in a pool can remain idle before it is disconnected; the age of a connection is based on the Least Recently Used (LRU) scheme.

String getNoWait() Retrieves whether the NOWAIT property is enabled. It returns a string of "true" or "false".

int getPoolSize()

OCI Connection Pooling 24-5

Connecting to an OCI Connection Pool

Retrieves the number of physical connections that are open. This should be used only as an estimate and for statistical analysis.

int getActiveSize() Retrieves the number of physical connections that are open and busy. This should be used only as an estimate and for statistical analysis.

boolean isPoolCreated() Retrieves whether the pool has been created. The pool is actually created when OracleOCIConnection(user, password, url, poolConfig) is called or when setUser, setPassword, and setURL has been done after calling OracleOCIConnection().

Connecting to an OCI Connection Pool

The OracleOCIConnectionPool class, through a getConnection method call, creates an instance of the OracleOCIConnection class. This instance represents a connection. Because the OracleOCIConnection class extends OracleConnection class, it has the functionality of this class too. Close the OracleOCIConnection objects once the user session is over, otherwise, they are closed when the pool instance is closed. There are two ways of calling getConnection:

OracleConnection getConnection() If you do not supply the user name and password, then the default user name and password used for the creation of the connection pool are used while creating the connection objects.

OracleConnection getConnection(String user, String password) Get a logical connection identified with the specified user name and password, which can be different from that used for pool creation.

The following code shows the signatures of the overloaded getConnection method:
public synchronized OracleConnection getConnection( ) throws SQLException /* * For getting a connection to the database. * * @param us Connection user-id * @param p Connection password * @return connection object */ public synchronized OracleConnection getConnection(String us, String p) throws SQLException

As an enhancement to OracleConnection, the following new method is added into OracleOCIConnection as a way to change password for the user:
void passwordChange (String user, String oldPassword, String newPassword)

24-6 Oracle Database JDBC Developer's Guide and Reference

Sample Code for OCI Connection Pooling

Sample Code for OCI Connection Pooling

The following code illustrates the use of OCI connection pooling in a sample application:
import import import import import import import import java.sql.DriverManager; java.sql.Connection; java.sql.PreparedStatement; java.sql.ResultSet; java.sql.SQLException; java.util.Properties; oracle.jdbc.OracleDriver; oracle.jdbc.pool.OracleOCIConnectionPool;

public class conPoolAppl extends Thread { public static final String query = "SELECT object_name FROM all_objects WHERE rownum < 300"; static public void main(String args[]) throws SQLException { int _maxCount = 10; Connection []conn = new Connection[_maxCount]; try { DriverManager.registerDriver(new OracleDriver()); String s = null; //System.getProperty ("JDBC_URL");

//String url = ( s == null ? "jdbc:oracle:oci8:@orcl" : s); String url = "jdbc:oracle:oci8:@orcl.rmmslang.com"; OracleOCIConnectionPool cpool = new OracleOCIConnectionPool("scott", "tiger", url, null); // Print out the default configuration for the OracleOCIConnectionPool System.out.println ("-- The default configuration for the OracleOCIConnectionPool --"); displayPoolConfig(cpool); //Set up the initial pool configuration Properties p1 = new Properties(); p1.put (OracleOCIConnectionPool.CONNPOOL_MIN_LIMIT, Integer.toString(1)); p1.put (OracleOCIConnectionPool.CONNPOOL_MAX_LIMIT, Integer.toString(_maxCount)); p1.put (OracleOCIConnectionPool.CONNPOOL_INCREMENT, Integer.toString(1)); // Enable the initial configuration cpool.setPoolConfig(p1); Thread []t = new Thread[_maxCount]; for (int i = 0; i < _maxCount; ++i) { conn[i] = cpool.getConnection("scott", "tiger"); if ( conn[i] == null ) { System.out.println("Unable to create connection."); return; } t[i] = new conPoolAppl (i, conn[i]); t[i].start (); //displayPoolConfig(cpool); OCI Connection Pooling 24-7

Sample Code for OCI Connection Pooling

} ((conPoolAppl)t[0]).startAllThreads (); try { Thread.sleep (200); } catch (Exception ea) {} displayPoolConfig(cpool); for (int i = 0; i < _maxCount; ++i) t[i].join (); } catch(Exception ex) { System.out.println("Error: " + ex); ex.printStackTrace (); return; } finally { for (int i = 0; i < _maxCount; ++i) if (conn[i] != null) conn[i].close (); } } //end of main private Connection m_conn; private static boolean m_startThread = false; private int m_threadId; public conPoolAppl (int i, Connection conn) { m_threadId = i; m_conn = conn; } public void startAllThreads () { m_startThread = true; } public void run () { while (!m_startThread) Thread.yield (); try { doQuery (m_conn); } catch (SQLException ea) { System.out.println ("*** Thread id: " + m_threadId); ea.printStackTrace (); } } // end of run private static void doQuery (Connection conn) throws SQLException { PreparedStatement pstmt = null; ResultSet rs = null;

24-8 Oracle Database JDBC Developer's Guide and Reference

Statement Handling and Caching

try { pstmt = conn.prepareStatement (query); rs = pstmt.executeQuery (); while (rs.next ()) { //System.out.println ("Object name: " +rs.getString (1)); } } catch (Exception ea) { System.out.println ("Error during execution: " +ea); ea.printStackTrace (); } finally { if (rs != null) rs.close (); if (pstmt != null) pstmt.close (); if (conn != null) conn.close (); } } // end of doQuery (Connection) // Display the current status of the OracleOCIConnectionPool private static void displayPoolConfig (OracleOCIConnectionPool cpool) throws SQLException { System.out.println (" Min poolsize Limit: " + cpool.getMinLimit()); System.out.println (" Max poolsize Limit: " + cpool.getMaxLimit()); /* System.out.println (" Connection Increment: " + cpool.getConnectionIncrement()); System.out.println (" NoWait: " + cpool.getNoWait()); System.out.println (" Timeout: " + cpool.getTimeout()); */ System.out.println (" PoolSize: " + cpool.getPoolSize()); System.out.println (" ActiveSize: " + cpool.getActiveSize()); } } // end of class conPoolAppl

Statement Handling and Caching

Statement caching is supported with OracleOCIConnectionPool. The caching improves performance by not having to open, parse, and close cursors. When OracleOCIConnection.prepareStatement ("a_SQL_query") is processed, the statement cache is searched for a statement that matches the SQL query. If a match is found, then we can reuse the Statement object instead of incurring the cost of creating another Statement object. The cache size can be dynamically increased or decreased. The default cache size is zero.
Note: The OracleStatement object created from OracleOCIConnection has the same behavior as one that is created from OracleConnection.

OCI Connection Pooling 24-9

JNDI and the OCI Connection Pool

JNDI and the OCI Connection Pool

The Java Naming and Directory Interface (JNDI) feature makes persistent the properties of Java object, therefore these properties can be used to construct a new instance of the object, such as cloning the object. The benefit is that the old object can be freed, and at a later time a new object with exactly the same properties can be created. The InitialContext.bind method makes persistent the properties, either on file or in a database, while the InitialContext.lookup method retrieves the properties from the persistent store and creates a new object with these properties. OracleOCIConnectionPool objects can be bound and looked up using the JNDI feature. No new interface calls in OracleOCIConnectionPool are necessary.

24-10 Oracle Database JDBC Developer's Guide and Reference

25
Oracle Advanced Queuing
Oracle Advanced Queuing (AQ) provides database-integrated message queuing functionality. It is built on top of Oracle Streams and leverages the functions of Oracle Database so that messages can be stored persistently, propagated between queues on different computers and databases, and transmitted using Oracle Net Services, HTTP, and HTTPS. Because Oracle AQ is implemented in database tables, all operational benefits of high availability, scalability, and reliability are also applicable to queue data. This chapter provides information about the Java interface to Oracle AQ.
See:

Oracle Streams Advanced Queuing User's Guide

This chapters covers the following topics:

Functionality and Framework AQ Asynchronous Event Notification Creating Messages Enqueuing Messages Dequeuing Messages Examples: Enqueuing and Dequeuing

Functionality and Framework

Oracle Database 11g Release 1 (11.1) provides a fast Java interface to Oracle AQ by introducing a new Java package, oracle.jdbc.aq. This package contains the following:

Classes AQDequeueOptions Specifies the options available for the dequeue operation AQEnqueueOptions Specifies the options available for the enqueue operation AQFactory Is a factory class for AQ AQNotificationEvent Is created whenever a new message is enqueued in a queue for which you have registered your interest

Oracle Advanced Queuing

25-1

AQ Asynchronous Event Notification

Interfaces AQAgent Used to represent and identify a user of the queue or a producer or consumer of the message AQMessage Represents a message that is enqueued or dequeued AQMessageProperties Contains message properties such as Correlation, Sender, Delay and Expiration, Recipients, and Priority and Ordering AQNotificationListener Is a listener interface for receiving AQ notification events AQNotificationRegistration Represents your interest in being notified when a new message is enqueued in a particular queue

These classes and interfaces enable you to access an existing queue, create messages, and enqueue and dequeue messages.
Note:

Oracle JDBC drivers do not provide any API to create a queue. Queues must be created through the DBMS_AQADM PL/SQL package. For more information about the APIs, refer to the JavaDoc.

See:

AQ Asynchronous Event Notification

A JDBC application can:

Register interest in notifications in the AQ namespace and receive notifications when an enqueue occurs Register interest in subscriptions to database events and receive notifications when the events are triggered

Registered clients are notified asynchronously when events are triggered or on an explicit AQ enqueue (or a new message is enqueued in a queue for which you have registered your interest). Clients do not need to be connected to a database. Example Example 251 illustrates the use of the new JDBC AQ interface which is new feature of the 11.1 JDBC thin driver. It shows how to enqueue and dequeue from a RAW type single and multi consumer queue. It also shows how AQ asynchronous notification works. In this example, the SCOTT user is connecting to the database. Therefore in the database you need to grant the following privileges to the user:
GRANT GRANT GRANT GRANT EXECUTE ON DBMS_AQ to SCOTT; EXECUTE ON DBMS_AQADM to SCOTT; AQ_ADMINISTRATOR_ROLE TO SCOTT; ADMINISTER DATABASE TRIGGER TO SCOTT; AQ asynchronous event notification example

Example 251

import java.sql.*; 25-2 Oracle Database JDBC Developer's Guide and Reference

AQ Asynchronous Event Notification

import java.util.Properties; import oracle.jdbc.*; import oracle.jdbc.aq.*; public class DemoAQRawQueue { static final String USERNAME= "scott"; static final String PASSWORD= "tiger"; static final String URL = "jdbc:oracle:thin:@(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=oracleserver.mydomain .com) (PORT=1521))"+"(CONNECT_DATA=(SERVICE_NAME=mydatabaseinstance)))"; public static final void main(String[] argv) { DemoAQRawQueue demo = new DemoAQRawQueue(); try { demo.run(); }catch(SQLException ex) { ex.printStackTrace(); } } void run() throws SQLException { OracleConnection connection = connect(); connection.setAutoCommit(false); cleanup(connection); setup(connection); // run the demo for single consumer queue: demoSingleConsumerQueue(connection); // run the demo for multi consumer queue: demoMultipleConsumerQueue(connection); connection.close(); } /** * Single consumer queue demo: * This method enqueues a dummy message into the RAW_SINGLE_QUEUE queue and dequeues it. * Then it registers for AQ notification on this same queue and enqueues a * message again. The AQ listener will be notified that a new message has arrived * in the queue and it will dequeue it. */ void demoSingleConsumerQueue(OracleConnection connection) throws SQLException { System.out.println("\n ============= Start single consumer queue demo ============= \n"); String queueType = "RAW"; String queueName = USERNAME+".RAW_SINGLE_QUEUE"; enqueueDummyMessage(connection,queueName,null); dequeue(connection,queueName,queueType,null); AQNotificationRegistration reg = registerForAQEvents( connection,queueName); DemoAQRawQueueListener single_li = new DemoAQRawQueueListener(

Oracle Advanced Queuing

25-3

AQ Asynchronous Event Notification

queueName,queueType); reg.addListener(single_li); enqueueDummyMessage(connection,queueName,null); connection.commit(); while(single_li.getEventsCount() < 1) { try { Thread.currentThread().sleep(1000); } catch (InterruptedException e) { } } single_li.closeConnection(); connection.unregisterAQNotification(reg); } /** * Multi consumer queue demo: * This method first registers for AQ notification upon the agent "BLUE". It * then enqueues a message for "RED" and "BLUE" */ void demoMultipleConsumerQueue(OracleConnection connection) throws SQLException { System.out.println("\n ============= Start multi consumer queue demo ============= \n"); String queueType = "RAW"; String queueName = USERNAME+".RAW_MULTIPLE_QUEUE"; AQNotificationRegistration reg = registerForAQEvents( connection,queueName+":BLUE"); DemoAQRawQueueListener multi_li = new DemoAQRawQueueListener( queueName,queueType); reg.addListener(multi_li); AQAgent[] recipients = new AQAgent[2]; recipients[0] = AQFactory.createAQAgent(); recipients[0].setName("BLUE"); recipients[1] = AQFactory.createAQAgent(); recipients[1].setName("RED"); enqueueDummyMessage(connection,queueName,recipients); connection.commit(); while(multi_li.getEventsCount() < 1) { try { Thread.currentThread().sleep(1000); } catch (InterruptedException e) { } } dequeue(connection,queueName,queueType,"RED"); multi_li.closeConnection(); connection.unregisterAQNotification(reg); }

/** * This method enqueues a dummy message in the queue specified by its name.

25-4 Oracle Database JDBC Developer's Guide and Reference

AQ Asynchronous Event Notification

*/ public void enqueueDummyMessage(OracleConnection conn, String queueName, AQAgent[] recipients) throws SQLException { System.out.println("----------- Enqueue start ------------"); // First create the message properties: AQMessageProperties msgprop = AQFactory.createAQMessageProperties(); msgprop.setCorrelation("mycorrelation"); msgprop.setExceptionQueue("MY_EXCEPTION_QUEUE"); // Specify an agent as the sender: AQAgent ag = AQFactory.createAQAgent(); ag.setName("MY_SENDER_AGENT_NAME"); ag.setAddress("MY_SENDER_AGENT_ADDRESS"); msgprop.setSender(ag); // handle multi consumer case: if(recipients != null) msgprop.setRecipientList(recipients); System.out.println(msgprop.toString()); // Create the actual AQMessage instance: AQMessage mesg = AQFactory.createAQMessage(msgprop); // and add a payload: byte[] rawPayload = new byte[500]; for(int i=0;i<rawPayload.length;i++) rawPayload[i] = 'b'; mesg.setPayload(new oracle.sql.RAW(rawPayload)); // We want to retrieve the message id after enqueue: AQEnqueueOptions opt = new AQEnqueueOptions(); opt.setRetrieveMessageId(true); // execute the actual enqueue operation: conn.enqueue(queueName,opt,mesg); byte[] mesgId = mesg.getMessageId(); if(mesgId != null) { String mesgIdStr = byteBufferToHexString(mesgId,20); System.out.println("Message ID from enqueue call: "+mesgIdStr); } System.out.println("----------- Enqueue done ------------"); } /** * This methods dequeues the next available message from the queue specified * by "queueName". */ public void dequeue(OracleConnection conn, String queueName, String queueType, String consumerName ) throws SQLException { System.out.println("----------- Dequeue start ------------"); AQDequeueOptions deqopt = new AQDequeueOptions();

Oracle Advanced Queuing

25-5

AQ Asynchronous Event Notification

deqopt.setRetrieveMessageId(true); if(consumerName != null) deqopt.setConsumerName(consumerName); // dequeue operation: AQMessage msg = conn.dequeue(queueName,deqopt,queueType); // print out the message that has been dequeued: byte[] payload = msg.getPayload(); byte[] msgId = msg.getMessageId(); if(msgId != null) { String mesgIdStr = byteBufferToHexString(msgId,20); System.out.println("ID of message dequeued = "+mesgIdStr); } AQMessageProperties msgProp = msg.getMessageProperties(); System.out.println(msgProp.toString()); String payloadStr = new String(payload,0,10); System.out.println("payload.length="+payload.length+", value="+payloadStr); System.out.println("----------- Dequeue done ------------"); }

public AQNotificationRegistration registerForAQEvents( OracleConnection conn, String queueName) throws SQLException { Properties globalOptions = new Properties(); String[] queueNameArr = new String[1]; queueNameArr[0] = queueName; Properties[] opt = new Properties[1]; opt[0] = new Properties(); opt[0].setProperty(OracleConnection.NTF_AQ_PAYLOAD,"true"); AQNotificationRegistration[] regArr = conn.registerAQNotification(queueNameArr,opt,globalOptions); AQNotificationRegistration reg = regArr[0]; return reg; } /** * */ private void setup(Connection conn) throws SQLException { doUpdateDatabase(conn, "BEGIN "+ "DBMS_AQADM.CREATE_QUEUE_TABLE( "+ " QUEUE_TABLE => '"+USERNAME+".RAW_SINGLE_QUEUE_TABLE', "+ " QUEUE_PAYLOAD_TYPE => 'RAW', "+ " COMPATIBLE => '10.0'); "+ "END; "); doUpdateDatabase(conn, "BEGIN "+ "DBMS_AQADM.CREATE_QUEUE( "+ " QUEUE_NAME => '"+USERNAME+".RAW_SINGLE_QUEUE', "+ " QUEUE_TABLE => '"+USERNAME+".RAW_SINGLE_QUEUE_TABLE'); "+ "END; "); doUpdateDatabase(conn,

25-6 Oracle Database JDBC Developer's Guide and Reference

AQ Asynchronous Event Notification

"BEGIN "+ " DBMS_AQADM.START_QUEUE('"+USERNAME+".RAW_SINGLE_QUEUE'); "+ "END; "); // create a multi consumer RAW queue: doUpdateDatabase(conn, "BEGIN "+ "DBMS_AQADM.CREATE_QUEUE_TABLE( "+ " QUEUE_TABLE => '"+USERNAME+".RAW_MULTIPLE_QUEUE_TABLE', "+ " QUEUE_PAYLOAD_TYPE => 'RAW', "+ " MULTIPLE_CONSUMERS => TRUE, "+ " COMPATIBLE => '10.0'); "+ "END; "); doUpdateDatabase(conn, "BEGIN "+ "DBMS_AQADM.CREATE_QUEUE( "+ " QUEUE_NAME => '"+USERNAME+".RAW_MULTIPLE_QUEUE', "+ " QUEUE_TABLE => '"+USERNAME+".RAW_MULTIPLE_QUEUE_TABLE'); "+ "END; "); doUpdateDatabase(conn, "BEGIN "+ " DBMS_AQADM.START_QUEUE('"+USERNAME+".RAW_MULTIPLE_QUEUE'); "+ "END; "); } void cleanup( Connection conn) { tryUpdateDatabase(conn, "BEGIN "+ " DBMS_AQADM.STOP_QUEUE('"+USERNAME+".RAW_SINGLE_QUEUE'); "+ "END; "); tryUpdateDatabase(conn, "BEGIN "+ " DBMS_AQADM.DROP_QUEUE_TABLE( "+ " QUEUE_TABLE => '"+USERNAME+".RAW_SINGLE_QUEUE_TABLE', "+ " FORCE => TRUE); "+ "END; "); tryUpdateDatabase(conn, "BEGIN "+ " DBMS_AQADM.STOP_QUEUE('"+USERNAME+".RAW_MULTIPLE_QUEUE'); "+ "END; "); tryUpdateDatabase(conn, "BEGIN "+ " DBMS_AQADM.DROP_QUEUE_TABLE( "+ " QUEUE_TABLE => '"+USERNAME+".RAW_MULTIPLE_QUEUE_TABLE', "+ " FORCE => TRUE); "+ "END; "); }

/** * Creates a connection the database. */ OracleConnection connect() throws SQLException {

Oracle Advanced Queuing

25-7

AQ Asynchronous Event Notification

OracleDriver dr = new OracleDriver(); Properties prop = new Properties(); prop.setProperty("user",DemoAQRawQueue.USERNAME); prop.setProperty("password",DemoAQRawQueue.PASSWORD); return (OracleConnection)dr.connect(DemoAQRawQueue.URL,prop); } /** * Utility method: executes a DML query but doesn't throw any exception if * an error occurs. */ private void tryUpdateDatabase(Connection conn,String sql) { Statement stmt = null; try { stmt = conn.createStatement(); stmt.executeUpdate(sql); }catch(SQLException sqlex) { System.out.println("Exception ("+sqlex.getMessage()+") while trying to execute \""+sql+"\""); } finally { if(stmt != null) { try { stmt.close(); }catch(SQLException exx){exx.printStackTrace();} } } } /** * Utility method: executes a DML query and throws an exception if * an error occurs. */ private void doUpdateDatabase(Connection conn,String sql) throws SQLException { Statement stmt = null; try { stmt = conn.createStatement(); stmt.executeUpdate(sql); } finally { if(stmt != null) { try { stmt.close(); }catch(SQLException exx){} } } } static final String byteBufferToHexString(byte[] buffer, int maxNbOfBytes)

25-8 Oracle Database JDBC Developer's Guide and Reference

AQ Asynchronous Event Notification

{ if(buffer == null) return null; int offset = 0; boolean isFirst = true; StringBuffer sb = new StringBuffer(); while(offset < buffer.length && offset < maxNbOfBytes) { if(!isFirst) sb.append(' '); else isFirst = false; String hexrep = Integer.toHexString((int)buffer[offset]&0xFF); if(hexrep.length() == 1) hexrep = "0"+hexrep; sb.append(hexrep); offset++; } String ret= sb.toString(); return ret; } } class DemoAQRawQueueListener implements AQNotificationListener { OracleConnection conn; String queueName; String typeName; int eventsCount = 0; public DemoAQRawQueueListener(String _queueName, String _typeName) throws SQLException { queueName = _queueName; typeName = _typeName; conn = (OracleConnection)DriverManager.getConnection (DemoAQRawQueue.URL, DemoAQRawQueue.USERNAME, DemoAQRawQueue.PASSWORD); } public void onAQNotification(AQNotificationEvent e) { System.out.println("\n----------- DemoAQRawQueueListener: got an event ----------- "); System.out.println(e.toString()); System.out.println("------------------------------------------------------"); System.out.println("----------- DemoAQRawQueueListener: Dequeue start -----------"); try { AQDequeueOptions deqopt = new AQDequeueOptions(); deqopt.setRetrieveMessageId(true); if(e.getConsumerName() != null) deqopt.setConsumerName(e.getConsumerName()); if((e.getMessageProperties()).getDeliveryMode() == AQMessageProperties.MESSAGE_BUFFERED) { deqopt.setDeliveryMode(AQDequeueOptions.DEQUEUE_BUFFERED); deqopt.setVisibility(AQDequeueOptions.DEQUEUE_IMMEDIATE); }

Oracle Advanced Queuing

25-9

Creating Messages

AQMessage msg = conn.dequeue(queueName,deqopt,typeName); byte[] msgId = msg.getMessageId(); if(msgId != null) { String mesgIdStr = DemoAQRawQueue.byteBufferToHexString(msgId,20); System.out.println("ID of message dequeued = "+mesgIdStr); } System.out.println(msg.getMessageProperties().toString()); byte[] payload = msg.getPayload(); if(typeName.equals("RAW")) { String payloadStr = new String(payload,0,10); System.out.println("payload.length="+payload.length+", value="+payloadStr); } System.out.println("----------- DemoAQRawQueueListener: Dequeue done -----------"); } catch(SQLException sqlex) { System.out.println(sqlex.getMessage()); } eventsCount++; } public int getEventsCount() { return eventsCount; } public void closeConnection() throws SQLException { conn.close(); } }

Creating Messages
Before you enqueue a message you need to create the message. An instance of a class implementing the AQMessage interface represents an AQ message. An AQ message contains properties (metadata) and a payload (data). To create an AQ message:
1. 2. 3. 4.

Create an instance of AQMessageProperties. Set the property attributes. Create the AQ message using the AQMessageProperties object. Set the payload.

AQ Message Properties The properties of the AQ message are represented by an instance of the AQMessageProperties interface. You can set or get the following message properties:

Dequeue Attempts Count: Specifies the number of attempts that have been made to dequeue the message. This property cannot be set. Correlation: Is an identifier supplied by the producer of the message at the time of enqueuing the message. Delay: Is the number of seconds for which the message is in the WAITING state. After the specified delay, the message is in the READY state and available for

25-10 Oracle Database JDBC Developer's Guide and Reference

Creating Messages

dequeuing. Dequeuing a message by using the message ID (msgid) overrides the delay specification.
Note:

Delay is not supported with buffered messaging.

Delivery Mode: Specifies whether the message is a buffered message or a persistent message. This property cannot be set. Enqueue Time: Specifies the time at which the message was enqueued. This value is determined by the system and cannot be set by the user. Exception Queue: Specifies the name of the queue into which the message is moved if it cannot be processed successfully. Messages are moved in two cases: The number of unsuccessful dequeue attempts has exceeded max_retries. The message has expired.

Expiration: Is the number of seconds during which the message is available for dequeuing, starting from when the message reaches the READY state. If the message is not dequeued before it expires, then it is moved to the exception queue in the EXPIRED state. Message State: Specifies the state of the message at the time of dequeuing the message. This property cannot be set. Previous Queue Message ID: Is the ID of the message in the last queue that generated the current message. When a message is propagated from one queue to another, this attribute identifies the ID of the queue from which it was last propagated. This property cannot be set. Priority: Specifies the priority of the message. It can be any integer including negative integers. Smaller the value, higher the priority. Recipient list: Is a list of AQAgent objects that represent the recipients. The default recipients are the queue subscribers. This parameter is valid only for multiconsumer queues. Sender: Is an identifier specified by the producer at the time of enqueuing the message. It is an instance of AQAgent. Transaction group: Specifies the transaction group of the message for transaction-grouped queues. It is set after a successful call to the dequeueArray method.

The following code snippet illustrates how to create an AQMessageProperties object and create an AQ message using it:
... // Create the message properties object AQMessageProperties msgprop = AQFactory.createAQMessageProperties(); // Set some properties (optional) msgprop.setCorrelation("mycorrelation"); msgprop.setExceptionQueue("MY_EXCEPTION_QUEUE"); msgprop.setExpiration(0); msgprop.setPriority(1); AQAgent ag = AQFactory.createAQAgent(); ag.setName("MY_SENDER_AGENT_NAME"); ag.setAddress("MY_SENDER_AGENT_ADDRESS"); msgprop.setSender(ag); msgprop.setRecipientList(recipients);

Oracle Advanced Queuing 25-11

Enqueuing Messages

// Create the message AQMessage mesg = AQFactory.createAQMessage(msgprop); ...

AQ Message Payload Depending on the type of the queue, the payload of the AQ message can be specified using the setPayload method of the AQMessage interface. The following code snippet illustrates how to set the payload:
... byte[] rawPayload = new byte[500]; for(int i=0;i<rawPayload.length;i++) rawPayload[i] = 'b'; mesg.setPayload(new oracle.sql.RAW(rawPayload)); ...

You can retrieve the payload of an AQ message using the getPayload method or the appropriate getXXXPayload method. These methods are defined in the AQMessage interface.

Enqueuing Messages
After you create a message and set the message properties and payload, you can enqueue the message using the enqueue method of the OracleConnection interface. Before you enqueue the message, you can specify some enqueue options. The AQEnqueueOptions class enables you to specify the following enqueue options:

Delivery mode: Specifies the delivery mode. Delivery mode can be set to either persistent (ENQUEUE_PERSISTENT) or buffered (ENQUEUE_BUFFERED). Retrieve Message ID: Specifies whether the message ID has to be retrieved from the server when the message has been enqueued. By default, the message ID is not retrieved. Transformation: Specifies a transformation that will be applied before enqueuing the message. The return type of the transformation function must match the type of the queue.
Note:

Transformations must be created in PL/SQL using DBMS_TRANSFORM.CREATE_TRANSFORMATION(...).

Visibility: Specifies the transactional behavior of the enqueue request. The default value for this option is ENQUEUE_ON_COMMIT. It indicates that the enqueue operation is part of the current transaction. ENQUEUE_IMMEDIATE indicates that the enqueue operation is an autonomous transaction, which commits at the end of the operation. For buffered messaging, you must use ENQUEUE_IMMEDIATE.

The following code snippet illustrates how to set the enqueue options and enqueue the message:
... // Set the enqueue options AQEnqueueOptions opt = new AQEnqueueOptions(); opt.setRetrieveMessageId(true); // Enqueue the message // conn is an instance of OracleConnection

25-12 Oracle Database JDBC Developer's Guide and Reference

Dequeuing Messages

// queueName is a String // mesg is an instance of AQMessage conn.enqueue(queueName, opt, mesg); ...

Dequeuing Messages
Enqueued messages can be dequeued using the dequeue method of the OracleConnection interface. Before you dequeue a message you must set the dequeue options. The AQDequeueOptions class enables you to specify the following dequeue options:

Condition: Specifies a conditional expression based on the message properties, the message data properties, and PL/SQL functions. A dequeue condition is specified as a Boolean expression using syntax similar to the WHERE clause of a SQL query. Consumer name: If specified, only the messages matching the consumer name are accessed.
Note:

If the queue is a single-consumer queue, do not set this option.

Correlation: Specifies a correlation criterion (or search criterion) for the dequeue operation. Delivery Filter: Specifies the type of message to be dequeued. You dequeue buffered messages only (DEQEUE_BUFFERED) or persistent messages only (DEQUEUE_PERSISTENT), which is the default, or both (DEQUEUE_PERSISTENT_OR_BUFFERED). Dequeue Message ID: Specifies the message identifier of the message to be dequeued. This can be used to dequeue a unique message whose ID is known. Dequeue mode: Specifies the locking behavior associated with the dequeue operation. It can take one of the following values: DequeueMode.BROWSE: Message is dequeued without acquiring any lock DequeueMode.LOCKED: Message is dequeued with a write lock that lasts for the duration of the transaction. DequeueMode.REMOVE: (default) Message is dequeued and deleted. The message can be retained in the queue based on the retention properties. DequeueMode.REMOVE_NO_DATA: Message is marked as updated or deleted.

Maximum Buffer Length: Specifies the maximum number of bytes that will be allocated when dequeuing a message from a RAW queue. The default maximum is DEFAULT_MAX_PAYLOAD_LENGTH but it can be changed to any other non zero value. If the buffer is not large enough to contain the entire message, then the exceeding bytes will be silently ignored. Navigation: Specifies the position of the message that will be retrieved. It can take one of the following values: NavigationOption.FIRST_MESSAGE: The first available message matching the search criteria is dequeued. NavigationOption.NEXT_MESSAGE: (default) The next available message matching the search criteria is dequeued. If the previous message belongs to a

Oracle Advanced Queuing 25-13

Dequeuing Messages

message group, then the next available message matching the search criteria in the message group is dequeued. NavigationOption.NEXT_TRANSACTION: Messages in the current transaction group are skipped, and the first message of the next transaction group is dequeued. This setting can be used only if message grouping is enabled for the queue.

Retrieve Message ID: Specifies whether the message identifier of the dequeued message needs to be retrieved. By default, it is not retrieved. Transformation: Specifies a transformation that will be applied after dequeuing the message. The source type of the transformation must match the type of the queue.
Note:

Transformations must be created in PL/SQL using DBMS_TRANSFORM.CREATE_TRANSFORMATION(...).

Visibility: Specifies whether the message is dequeued as part of the current transaction. It can take one of the following values: VisibilityOption.ON_COMMIT: (default) The dequeue operation is part of the current transaction. VisibilityOption.IMMEDIATE: The dequeue operation is an autonomous transaction that commits at the end of the operation.
Note: The Visibility option is ignored in the DequeueMode.BROWSE dequeue mode. If the delivery filter is DEQUEUE_BUFFERED or DEQUEUE_PERSISTENT_OR_BUFFERED, then this option must be set to VisibilityOption.IMMEDIATE.

Wait: Specifies the wait time for the dequeue operation, if none of the messages match the search criteria. The default value is DEQUEUE_WAIT_FOREVER indicating that the operation waits forever. If set to DEQUEUE_NO_WAIT, then the operation does not wait. If a number is specified, then the dequeue operation waits for the specified number of seconds.
Note:

If you use DEQUEUE_WAIT_FOREVER, then the dequeue operation will not return until a message which matches the search criterion is available in the queue. However, you can interrupt the dequeue operation by calling the cancel method on the OracleConnection object.

The following code snippet illustrates how to set the dequeue options and dequeue the message:
... // Set the dequeue options AQDequeueOptions deqopt = new AQDequeueOptions(); deqopt.setRetrieveMessageId(true); deqopt.setConsumerName(consumerName); // Dequeue the message // conn is an OracleConnection object // queueName is a String identifying the queue // queueType is a String specifying the type of the queue, such as RAW

25-14 Oracle Database JDBC Developer's Guide and Reference

Examples: Enqueuing and Dequeuing

AQMessage msg = conn.dequeue(queueName,deqopt,queueType);

Examples: Enqueuing and Dequeuing

This section provides a few examples that illustrate how to enqueue and dequeue messages. Example 252 illustrates how to enqueue a message, and Example 253 illustrates how to dequeue a message.
Example 252 Enqueuing a Single Message

This example illustrates how to obtain access to a queue, create a message, and enqueue it.
// Get access to the queue // conn is an instance of oracle.jdbc.OracleConnection oracle.jdbc.aq.AQQueue queue = conn.getAQQueue("SCOTT.MY_QUEUE","RAW"); // Create the message properties object oracle.jdbc.aq.AQMessageProperties msgprop = oracle.jdbc.aq.AQMessageProperties.createAQMessageProperties(); // Set some properties (optional) msgprop.setPriority(1); msgprop.setExceptionQueue("EXCEPTION_QUEUE"); msgprop.setExpiration(0); AQAgent agent = AQAgent.createAQAgent("AGENTNAME"); ag2.setAddress("AGENTADDRESS"); msgprop.setSender(agent); // Create the message AQMessage mesg = AQMessage.createAQMessage(msgprop); // Set the payload // where buffer contains the RAW payload: mesg.setRawPayload(new AQRawPayload(buffer)); // Set the enqueue options AQEnqueueOptions options = new AQEnqueueOptions(); //Set the enqueue options using the setXXX methods of AQEnqueueOptions. // Enqueue the message queue.enqueue(options, mesg);

Example 253

Dequeuing a Single Message

This example illustrates how to obtain access to a queue, set the dequeue options, and dequeue the message.
// Get access to the queue // conn is an instance of oracle.jdbc.OracleConnection oracle.jdbc.aq.AQQueue queue = conn.getAQQueue ("SCOTT.MY_QUEUE","RAW"); // Set the dequeue options AQDequeueOptions options = new AQDequeueOptions(); options.setDeliveryMode(AQDequeueOptions.BUFFERED); // Dequeue the message AQMessage mesg = queue.dequeue(options); Oracle Advanced Queuing 25-15

Examples: Enqueuing and Dequeuing

25-16 Oracle Database JDBC Developer's Guide and Reference

Part VI
High Availability
This section provides information about the high-availability features of Oracle Database 10g. It discusses the Fast Connection Failover and Transparent Application Failover (TAF) features Part VI contains the following chapters:

Chapter 26, "Fast Connection Failover" Chapter 27, "Transparent Application Failover"

26
Fast Connection Failover
The Fast Connection Failover mechanism depends on the implicit connection cache feature. As a result, for Fast Connection Failover to be available, implicit connection caching must be enabled. This chapter is divided into the following sections:

Overview of Fast Connection Failover Using Fast Connection Failover Understanding Fast Connection Failover Comparison of Fast Connection Failover and TAF
See Also:

Chapter 21, "Implicit Connection Caching"

Overview of Fast Connection Failover

Fast Connection Failover offers a driver-independent way for your Java Database Connectivity (JDBC) application to take advantage of the connection failover facilities offered by Oracle Database 10g. The advantages of Fast Connection Failover include:

Driver independence Fast Connection Failover supports both the JDBC Thin and JDBC Oracle Call Interface (OCI) drivers.

Integration with implicit connection cache The two features work together synergistically to improve application performance and high availability.

Integration with Oracle Real Application Clusters (RAC) This provides superior Real Application Clusters/High Availability event notification mechanisms.

Easy integration with application code You only need to enable Fast Connection Failover and no further configuration is required.

Fast Connection Failover Features When enabled, Fast Connection Failover provides:

Rapid detection and cleanup of invalid cached connections, that is, DOWN event processing Load balancing of available connections, that is, UP event processing
Fast Connection Failover 26-1

Using Fast Connection Failover

Run-time work request distribution to all active RAC instances

Using Fast Connection Failover

Applications manage Fast Connection Failover through DataSource instances. This section covers the following topics:

Fast Connection Failover Prerequisites Configuring ONS For Fast Connection Failover Enabling Fast Connection Failover Querying Fast Connection Failover Status

Fast Connection Failover Prerequisites

Fast Connection Failover is available under the following circumstances:

The implicit connection cache is enabled Fast Connection Failover works in conjunction with the JDBC connection caching mechanism. This helps applications manage connections to ensure high availability.

The application uses service names to connect to the database The application cannot use service identifiers (SIDs).

The underlying database has Oracle Database 10g Real Application Clusters capability If failover events are not propagated, then connection failover cannot occur.

Oracle Notification Service (ONS) is configured and available on the node where JDBC is running JDBC depends on ONS to propagate database events and notify JDBC of them.

The Java Virtual Machine (JVM) in which your JDBC instance is running must have oracle.ons.oraclehome set to point to your ORACLE_HOME.

Configuring ONS For Fast Connection Failover

In order for Fast Connection Failover to work, you must configure ONS correctly. ONS is shipped as part of Oracle Database 10g. This section covers the following topics:

ONS Configuration File Client-Side ONS Configuration Server-Side ONS Configuration Using racgons Remote ONS Subscription

ONS Configuration File

ONS configuration is controlled by the ONS configuration file, ORACLE_HOME/opmn/conf/ons.config. This file tells the ONS daemon details about how it should behave and who it should talk to. Configuration information within ons.config is defined in simple name/value pairs. There are three values that should always be configured within ons.config. The first is localport, the
26-2 Oracle Database JDBC Developer's Guide and Reference

Using Fast Connection Failover

port that ONS binds to on the localhost interface to talk to local clients. An example of the localport configuration is:
localport=4100

The second value is remoteport, the port that ONS binds to on all interfaces for talking to other ONS daemons. An example of the remoteport configuration is:
remoteport=4200

The third value specifies nodes, a list of other ONS daemons to talk to. Node values are given as a comma-delimited list of either host names or IP addresses plus ports. Note that the port value that is given is the remote port that each ONS instance is listening on. In order to maintain an identical file on all nodes, the host:port of the current ONS node can also be listed in the nodes list. It will be ignored when reading the list. The nodes listed in the nodes line correspond to the individual nodes in the RAC instance. Listing the nodes ensures that the middle-tier node can communicate with the RAC nodes. At least one mid-tier node and one node in the RAC instance must be configured to see one another. As long as one node on each side is aware of the other, all nodes are visible. You need not list every single cluster and middle-tier node in the ONS config file of each RAC node. In particular, if one ONS config file cluster node is aware of the middle tier, then all nodes in the cluster are aware of it. An example of the nodes configuration is:
nodes=myhost.example.com:4200,123.123.123.123:4200

There are also several optional values that can be provided in ons.config. The first optional value is a loglevel. This specifies the level of messages that should be logged by ONS. This value is an integer that ranges from 1, which indicates least messages logged, to 9, which indicates most messages logged. The default value is 3. An example is:
loglevel=3

The second optional value is a logfile name. This specifies a log file that ONS should use for logging messages. The default value for logfile is $ORACLE_HOME/opmn/logs/ons.log. An example is:
logfile=/private/oraclehome/opmn/logs/myons.log

The third optional value is a walletfile name. A wallet file is used by the Oracle Secure Sockets Layer (SSL) to store SSL certificates. If a wallet file is specified to ONS, it will use SSL when communicating with other ONS instances and require SSL certificate authentication from all ONS instances that try to connect to it. This means that if you want to turn on SSL for one ONS instance, then you must turn it on for all instances that are connected. This value should point to the directory where your ewallet.p12 file is located. An example is:
walletfile=/private/oraclehome/opmn/conf/ssl.wlt/default

One optional value is reserved for use on the server-side. useocr=on is used to tell ONS to store all RAC nodes and port numbers in Oracle Cluster Registry (OCR) instead of in the ONS configuration file. Do not use this option on the client-side. The ons.config file allows blank lines and comments on lines that begin with #.

Fast Connection Failover 26-3

Using Fast Connection Failover

Client-Side ONS Configuration

You can access the client-side ONS through ORACLE_HOME/opmn. On the client-side, there are two ways to set up ONS:

Remote ONS configuration

See Also:

"Remote ONS Subscription" on page 26-5

ONS daemon on the client-side

Example 261 illustrates how a sample configuration file may look like.
Example 261 ons.config file

# This is an example ons.config file # # The first three values are required localport=4100 remoteport=4200 nodes=racnode1.example.com:4200,racnode2.example.com:4200

After configuring ONS, you start the ONS daemon with the onsctl command. It is the user's responsibility to make sure that an ONS daemon is running at all times. Using the onsctl Command After configuring, use ORACLE_HOME/opmn/bin/onsctl to start, stop, reconfigure, and monitor the ONS daemon. Table 261 is a summary of the commands that onsctl supports.
Table 261 Command start stop ping reconfig onsctl commands Effect Starts the ONS daemon Stops the ONS daemon Verifies whether the ONS daemon is running Triggers a reload of the ONS configuration without shutting down the ONS daemon Prints a help summary message for onsctl Prints a detailed help message for onsctl Output onsctl: ons started onsctl: shutting down ons daemon... ons is running ...

help detailed

Server-Side ONS Configuration Using racgons

You can access the server-side ONS through ORA_CRS_HOME/opmn. You configure the server-side by using racgons to add the middle-tier node information to OCR. This command is found in ORA_CRS_HOME/bin/racgons. Before using racgons, you must edit ons.config to set useocr=on. The middle-tier nodes should be configured in OCR, so that all nodes share the configuration, and no matter which RAC nodes are up they can communicate to the mid-tier. When running on a cluster, always configure the ONS hosts and ports not by using the ONS configuration files but using racgons. The racgons command stores the ONS hosts and ports in OCR, where every node can see it. That way, you don't

26-4 Oracle Database JDBC Developer's Guide and Reference

Using Fast Connection Failover

need to edit a file on every node to change the configuration, just run a single command on one of the cluster nodes. The racogns command enables you to specify hosts and ports on one node, then propagate your changes among all nodes in a cluster. The command takes two forms:
racgons add_config hostname:port [hostname:port] [hostname:port] ... racgons remove_config hostname[:port] [hostname:port] [hostname:port] ...

The add_config version adds the listed host name(s), the remove_config version removes them. Both commands propagate the changes among all instances in a cluster. If multiple port numbers are configured for a host, the specified port number is removed from hostname. If only hostname is specified, all port numbers for that host are removed.
See Also: Oracle Real Application Clusters Administration and Deployment Guide

Other Uses of racgons You should run racgons whenever you add a new node to the cluster.

Remote ONS Subscription

The advantages of remote ONS subscription are:

Support for an All Java mid-tier stack No ONS daemon needed on the client computer and, therefore, no need to manage this process Simple configuration using the DataSource property.

When using remote ONS subscription for Fast Connection Failover, the application invokes the following method on an OracleDataSource instance:
setONSConfiguration(String remoteONSConfig)

The remoteONSConfig parameter is a list of name/value pairs of the form name=value that are separated by a new line character (\n). name can be one of nodes, walletfile, or walletpassword. This parameter should specify at least the nodes ONS configuration attribute, which is a list of host:port pairs separated by comma (,). The hosts and ports denote the remote ONS daemons available on the RAC nodes.
See Also:

"ONS Configuration File" on page 26-2

SSL could be used in communicating with the ONS daemons when the walletfile attribute is specified as an Oracle wallet file. In such cases, if the walletpassword attribute is not specified, single sign-on (SSO) would be assumed. Following are a few examples, assuming ods is an OracleDataSource instance:
ods.setONSConfiguration("nodes=racnode1.example.com:4200,racnode2.example.com:4200 "); ods.setONSConfiguration("nodes=racnode1:4200,racnode2:4200\nwalletfile=/mydir/Wall et\nwalletpassword=mypasswd"); ods.setONSConfiguration("nodes=racnode1:4200,racnode2:4200\nwalletfile=/mydir/conf /Wallet");

Fast Connection Failover 26-5

Using Fast Connection Failover

Note: The ons.jar must be in the CLASSPATH on the client. In the case of Oracle Application Server, ONS is embedded in OPMN, as before, and JDBC Fast Connection Failover continues to work as before.

Enabling Fast Connection Failover

An application enables Fast Connection Failover by calling setFastConnectionFailoverEnabled(true) on a DataSource instance, before retrieving any connections from that instance. You cannot enable Fast Connection Failover when reinitializing a connection cache. You must enable it before using the OracleDataSource instance. Example 262 illustrates how to enable Fast Connection Failover.
Note:

After a cache is Fast Connection Failover-enabled, you cannot disable Fast Connection Failover during the lifetime of that cache.

To enable Fast Connection Failover, you must:

Configure and start ONS. If ONS is not correctly set up, then implicit connection cache creation fails and an ONSException is thrown at the first getConnection request. Set the FastConnectionFailoverEnabled property before making the first getConnection request to an OracleDataSource. When Fast Connection Failover is enabled, the failover applies to all connections in the connection cache. If your application explicitly creates a connection cache using the Connection Cache Manager, then you must first set FastConnectionFailoverEnabled before retrieving any connections. Use a service name rather than an SID when setting the OracleDataSource url property.
Enabling Fast Connection Failover

Example 262

// declare datasource ods.setUrl( "jdbc:oracle:oci:@(DESCRIPTION= (ADDRESS=(PROTOCOL=TCP)(HOST=cluster_alias) (PORT=1521)) (CONNECT_DATA=(SERVICE_NAME=service_name)))"); ods.setUser("scott"); ods.setConnectionCachingEnabled(true); ods.setFastConnectionFailoverEnabled(true): ctx.bind("myDS",ods); ds=(OracleDataSource) ctx.lookup("MyDS"); try { ds.getConnection(); // transparently creates and accesses cache catch (SQLException SE { } } ...

26-6 Oracle Database JDBC Developer's Guide and Reference

Understanding Fast Connection Failover

Querying Fast Connection Failover Status

An application determines whether Fast Connection Failover is enabled by calling OracleDataSource.getFastConnectionFailoverEnabled, which returns true if failover is enabled, false otherwise.

Understanding Fast Connection Failover

After Fast Connection Failover is enabled, the mechanism is automatic; no application intervention is needed. This section discusses how a connection failover is presented to an application and what steps the application takes to recover. This section covers the following topics:

What The Application Sees How It Works

What The Application Sees

When a RAC service failure is propagated to the JDBC application, the database has already rolled back the local transaction. The cache manager then cleans up all invalid connections. When an application holding an invalid connection tries to do work through that connection, it is possible to receive SQLException, ORA-17008, Closed Connection. When an application receives a Closed Connection error message, it should:
1. 2.

Retry the connection request. This is essential, because the old connection is no longer open. Replay the transaction. All work done before the connection was closed has been lost.
Note: The application should not try to roll back the transaction. The transaction was already rolled back in the database by the time the application received the exception.

How It Works
Under Fast Connection Failover, each connection in the cache maintains a mapping to a service, instance, database, and host name. When a database generates a RAC event, that event is forwarded to the JVM in which JDBC is running. A daemon thread inside the JVM receives the RAC event and passes it on to the Connection Cache Manager. The Connection Cache Manager then throws SQL exceptions to the applications affected by the RAC event. A typical failover scenario may work like this:
1. 2. 3.

A database instance fails, leaving several stale connections in the cache. The RAC mechanism in the database generates a RAC event which is sent to the JVM containing JDBC. The daemon thread inside the JVM finds all the connections affected by the RAC event, notifies them of the closed connection through SQL exceptions, and rolls back any open transactions. Each individual connection receives a SQL exception and must retry.

4.

Fast Connection Failover 26-7

Comparison of Fast Connection Failover and TAF

Comparison of Fast Connection Failover and TAF

Fast Connection Failover differs from Transparent Application Failover (TAF) in the following ways:

Application-level connection retries Fast Connection Failover supports application-level connection retries. This gives the application control of responding to connection failovers. The application can choose whether to retry the connection or to rethrow the exception. TAF supports connection retries only at the OCI/Net layer.

Integration with the implicit connection cache Fast Connection Failover is well-integrated with the implicit connection cache, which allows the connection cache manager to manage the cache for high availability. For example, failed connections are automatically invalidated in the cache. TAF works at the network level on a per-connection basis, which means that the connection cache cannot be notified of failures.

Event-based Fast Connection Failover is based on the RAC event mechanism. This means that Fast Connection Failover is efficient and detects failures quickly for both active and inactive connections.

Load-balancing support Fast Connection Failover supports UP event load balancing of connections and run-time work request distribution across active RAC instances.
Note:

Oracle recommends not to use TAF and Fast Connection Failover in the same application.

26-8 Oracle Database JDBC Developer's Guide and Reference

27
Transparent Application Failover
This chapter contains the following sections:

Overview Failover Type Events TAF Callbacks Java TAF Callback Interface

Overview
Transparent Application Failover (TAF) is a feature of the Java Database Connectivity (JDBC) Oracle Call Interface (OCI) driver. It enables the application to automatically reconnect to a database, if the database instance to which the connection is made fails. In this case, the active transactions roll back. When an instance to which a connection is established fails or is shutdown, the connection on the client-side becomes stale and would throw exceptions to the caller trying to use it. TAF enables the application to transparently reconnect to a preconfigured secondary instance creating a fresh connection, but identical to the connection that was established on the first original instance. That is, the connection properties are the same as that of the earlier connection. This is true regardless of how the connection was lost.
Note:

TAF is always active and does not have to be set. TAF does not work with the OCI Connection Pool.

Failover Type Events

The following are possible failover events in the OracleOCIFailover interface:

FO_SESSION Is equivalent to FAILOVER_MODE=SESSION in the tnsnames.ora file CONNECT_DATA flags. This means that only the user session is authenticated again on the server-side, while open cursors in the OCI application need to be reprocessed.

FO_SELECT Is equivalent to FAILOVER_MODE=SELECT in tnsnames.ora file CONNECT_DATA flags. This means that not only the user session is
Transparent Application Failover 27-1

TAF Callbacks

re-authenticated on the server-side, but open cursors in the OCI can continue fetching. This implies that the client-side logic maintains fetch-state of each open cursor.

FO_NONE Is equivalent to FAILOVER_MODE=NONE in the tnsnames.ora file CONNECT_DATA flags. This is the default, in which no failover functionality is used. This can also be explicitly specified to prevent failover from happening. Additionally, FO_TYPE_UNKNOWN implies that a bad failover type was returned from the OCI driver.

FO_BEGIN Indicates that failover has detected a lost connection and failover is starting.

FO_END Indicates successful completion of failover.

FO_ABORT Indicates that failover was unsuccessful and there is no option of retrying.

FO_REAUTH Indicates that a user handle has been re-authenticated.

FO_ERROR Indicates that failover was temporarily unsuccessful, but it gives the application the opportunity to handle the error and retry failover. The usual method of error handling is to issue the sleep method and retry by returning the value FO_RETRY.

FO_RETRY Indicates that the application should retry failover.

FO_EVENT_UNKNOWN Indicates a bad failover event.

TAF Callbacks
TAF callbacks are used in the event of the failure of one database connection, and failover to another database connection. TAF callbacks are callbacks that are registered in case of failover. The callback is called during the failover to notify the JDBC application of events generated. The application also has some control of failover.
Note:

The callback setting is optional.

Java TAF Callback Interface

The OracleOCIFailover interface includes the callbackFn method, supporting the following types and events:
public interface OracleOCIFailover{ // Possible Failover Types public static final int FO_SESSION = 1; public static final int FO_SELECT = 2; public static final int FO_NONE = 3;

27-2 Oracle Database JDBC Developer's Guide and Reference

Java TAF Callback Interface

public static final int; // Possible Failover events registered with callback public static final int FO_BEGIN = 1; public static final int FO_END = 2; public static final int FO_ABORT = 3; public static final int FO_REAUTH = 4; public static final int FO_ERROR = 5; public static final int FO_RETRY = 6; public static final int FO_EVENT_UNKNOWN = 7; public int callbackFn (Connection conn, Object ctxt, // ANy thing the user wants to save int type, // One of the possible Failover Types int event ); // One of the possible Failover Events

Handling the FO_ERROR Event In case of an error while failing-over to a new connection, the JDBC application is able to retry failover. Typically, the application sleeps for a while and then it retries, either indefinitely or for a limited amount of time, by having the callback return FO_RETRY. Handling the FO_ABORT Event Callback registered should return the FO_ABORT event if the FO_ERROR event is passed to it.

Transparent Application Failover 27-3

Java TAF Callback Interface

27-4 Oracle Database JDBC Developer's Guide and Reference

Part VII
Transaction Management
This part provides information about transaction management in Oracle Java Database Connectivity (JDBC). It includes a chapter that discusses the Oracle JDBC implementation of distributed transactions. Part VII contains the following chapter:

Chapter 28, "Distributed Transactions"

28
Distributed Transactions
This chapter discusses the Oracle Java Database Connectivity (JDBC) implementation of distributed transactions. These are multiphased transactions, often using multiple databases, which must be committed in a coordinated way. There is also related discussion of XA, which is a general standard, and not specific to Java, for distributed transactions. The following topics are discussed:

Overview XA Components Error Handling and Optimizations Implementing a Distributed Transaction Native-XA in Oracle JDBC Drivers
Note:

This chapter discusses features of the JDBC 2.0 Optional Package, formerly known as the JDBC 2.0 Standard Extension application programming interface (API), which is available through the javax packages from Sun Microsystems.

For further introductory and general information about distributed transactions, refer to the Sun Microsystems specifications for the JDBC 2.0 Optional Package and the Java Transaction API (JTA).

Overview
A distributed transaction, sometimes referred to as a global transaction, is a set of two or more related transactions that must be managed in a coordinated way. The transactions that constitute a distributed transaction might be in the same database, but more typically are in different databases and often in different locations. Each individual transaction of a distributed transaction is referred to as a transaction branch. For example, a distributed transaction might consist of money being transferred from an account in one bank to an account in another bank. You would not want either transaction committed without assurance that both will complete successfully. In the JDBC 2.0 extension API, distributed transaction functionality is built on top of connection pooling functionality. This distributed transaction functionality is also built upon the open XA standard for distributed transactions. XA is part of the X/Open standard and is not specific to Java.

Distributed Transactions 28-1

Overview

JDBC is used to connect to database resources. However, to include all changes to multiple databases within a transaction, you must use the JDBC connections within a JTA global transaction. The process of including database SQL updates within a transaction is referred to as enlisting a database resource. The section covers the following topics:

Distributed Transaction Components and Scenarios Distributed Transaction Concepts Switching Between Global and Local Transactions Oracle XA Packages

Distributed Transaction Components and Scenarios

In reading the remainder of the distributed transactions section, it will be helpful to keep the following points in mind:

A distributed transaction system typically relies on an external transaction manager, such as a software component that implements standard JTA functionality, to coordinate the individual transactions. Many vendors offer XA-compliant JTA modules, including Oracle, which includes JTA in Oracle9i Application Server and Oracle Application Server 10g.

XA functionality is usually isolated from a client application, being implemented instead in a middle-tier environment, such as an application server. In many scenarios, the application server and transaction manager will be together on the middle tier, possibly together with some of the application code as well.

Discussion throughout this section is intended mostly for middle-tier developers. The term resource manager is often used in discussing distributed transactions. A resource manager is simply an entity that manages data or some other kind of resource. Wherever the term is used in this chapter, it refers to a database.
Note: Using JTA functionality requires jta.jar to be in the CLASSPATH environment variable. This file is located at ORACLE_HOME/jlib. Oracle includes this file with the JDBC product. You can also obtain it from the Sun Microsystems Web site, but it is advisable to use the version from Oracle, because that has been tested with the Oracle drivers.

Distributed Transaction Concepts

When you use XA functionality, the transaction manager uses XA resource instances to prepare and coordinate each transaction branch and then to commit or roll back all transaction branches appropriately. XA functionality includes the following key components:

XA data sources These are extensions of connection pool data sources and other data sources, and similar in concept and functionality. There will be one XA data source instance for each resource manager that will be used in the distributed transaction. You will typically create XA data source instances in your middle-tier software.

28-2 Oracle Database JDBC Developer's Guide and Reference

Overview

XA data sources produce XA connections.

XA connections These are extensions of pooled connections and similar in concept and functionality. An XA connection encapsulates a physical database connection. Individual connection instances are temporary handles to these physical connections. An XA connection instance corresponds to a single Oracle session, although the session can be used in sequence by multiple logical connection instances, as with pooled connection instances. You will typically get an XA connection instance from an XA data source instance in your middle-tier software. You can get multiple XA connection instances from a single XA data source instance if the distributed transaction will involve multiple sessions in the same database. XA connections produce OracleXAResource instances and JDBC connection instances.

XA resources These are used by a transaction manager in coordinating the transaction branches of a distributed transaction. You will get one OracleXAResource instance from each XA connection instance, typically in your middle-tier software. There is a one-to-one correlation between OracleXAResource instances and XA connection instances. Equivalently, there is a one-to-one correlation between OracleXAResource instances and Oracle sessions. In a typical scenario, the middle-tier component will hand off OracleXAResource instances to the transaction manager, for use in coordinating distributed transactions. Because each OracleXAResource instance corresponds to a single Oracle session, there can be only a single active transaction branch associated with an OracleXAResource instance at any given time. However, there can be additional suspended transaction branches. Each OracleXAResource instance has the functionality to start, end, prepare, commit, or roll back the operations of the transaction branch running in the session with which the OracleXAResource instance is associated. The prepare step is the first step of a two-phase commit operation. The transaction manager will issue a prepare to each OracleXAResource instance. Once the transaction manager sees that the operations of each transaction branch have prepared successfully, it will issue a COMMIT to each OracleXAResource instance to commit all the changes.

Transaction IDs These are used to identify transaction branches. Each ID includes a transaction branch ID component and a distributed transaction ID component. This is how a branch is associated with a distributed transaction. All OracleXAResource instances associated with a given distributed transaction would have a transaction ID that includes the same distributed transaction ID component.

OracleXAResource.ORATRANSLOOSE Start a loosely coupled transaction with transaction ID xid.

Distributed Transactions 28-3

Overview

Switching Between Global and Local Transactions

As of JDBC 3.0, applications can share connections between local and global transactions. Applications can also switch connections between local transactions and global transactions. A connection is always in one of the following modes:

NO_TXN No transaction is actively using this connection.

LOCAL_TXN A local transaction with auto-commit turned off or disabled is actively using this connection.

GLOBAL_TXN A global transaction is actively using this connection.

Each connection switches automatically between these modes depending on the operations carried out on the connection. A connection is always in NO_TXN mode when it is instantiated.
Note:

The modes are maintained internally by the JDBC drivers in association with Oracle Database.

Table 281 describes the connection mode transition rules.

Table 281 Connection Mode Transitions Switches To NO_TXN When NA Switches to LOCAL_TXN When Auto-commit mode is false and an Oracle data manipulation language (DML) statement is run. NA Switches To GLOBAL_TXN When start is called on an XAResource obtained from the XAconnection that provided this connection. NEVER

Current Mode NO_TXN

LOCAL_TXN

Any of the following happens:

An Oracle data definition language (DDL) statement is run commit is called rollback is called, but without parameters NA

GLOBAL_TXN

NEVER Within a global transaction open on this connection, end is called on an XAResource obtained from the XAconnection that provided this connection.

28-4 Oracle Database JDBC Developer's Guide and Reference

XA Components

If none of these rules are applicable, then the mode does not change. Mode Restrictions On Operations The current connection mode restricts which operations are valid within a transaction.

In the LOCAL_TXN mode, applications must not call start, prepare, commit, rollback, forget, or end on an XAResource. Doing so causes an XAException to be thrown. In the GLOBAL_TXN mode, applications must not call commit, rollback, rollback(Savepoint), setAutoCommit(true), or setSavepoint on a java.sql.Connection, and must not call OracleSetSavepoint or oracleRollback on an oracle.jdbc.OracleConnection. Doing so causes an SQLException to be thrown.
Note:

This mode-restriction error checking is in addition to the standard error checking on the transaction and savepoint APIs.

Oracle XA Packages
Oracle supplies the following three packages that have classes to implement distributed transaction functionality according to the XA standard:

oracle.jdbc.xa oracle.jdbc.xa.client oracle.jdbc.xa.server

Classes for XA data sources, XA connections, and XA resources are in both the client package and the server package. An abstract class for each is in the top-level package. The OracleXid and OracleXAException classes are in the top-level oracle.jdbc.xa package, because their functionality does not depend on where the code is running. In middle-tier scenarios, you will import OracleXid, OracleXAException, and the oracle.jdbc.xa.client package. If you intend your XA code to run in the target Oracle Database, however, you will import the oracle.jdbc.xa.server package instead of the client package. If code that will run inside a target database must also access remote databases, then do not import either package. Instead, you must fully qualify the names of any classes that you use from the client package to access a remote database or from the server package to access the local database. Class names are duplicated between these packages.

XA Components
This section discusses the XA components, that is, standard XA interfaces specified in the JDBC 2.0 Optional Package, and the Oracle classes that implement them. The following topics are covered:

XADatasource Interface and Oracle Implementation XAConnection Interface and Oracle Implementation XAResource Interface and Oracle Implementation OracleXAResource Method Functionality and Input Parameters

Distributed Transactions 28-5

XA Components

Xid Interface and Oracle Implementation

XADatasource Interface and Oracle Implementation

The javax.sql.XADataSource interface outlines standard functionality of XA data sources, which are factories for XA connections. The overloaded getXAConnection method returns an XA connection instance and optionally takes a user name and password as input:
public interface XADataSource { XAConnection getXAConnection() throws SQLException; XAConnection getXAConnection(String user, String password) throws SQLException; ... }

Oracle JDBC implements the XADataSource interface with the OracleXADataSource class, located both in the oracle.jdbc.xa.client package and the oracle.jdbc.xa.server package. The OracleXADataSource classes also extend the OracleConnectionPoolDataSource class, which extends the OracleDataSource class, and therefore, include all the connection properties. The getXAConnection methods of the OracleXADataSource class returns the Oracle implementation of XA connection instances, which are OracleXAConnection instances.
Note:

You can register XA data sources in Java Naming Directory and Interface (JNDI) using the same naming conventions as discussed previously for non-pooling data sources.

See Also: For information about Fast Connection Failover, see Chapter 26, "Fast Connection Failover".

XAConnection Interface and Oracle Implementation

An XA connection instance, as with a pooled connection instance, encapsulates a physical connection to a database. This would be the database specified in the connection properties of the XA data source instance that produced the XA connection instance. Each XA connection instance also has the facility to produce the OracleXAResource instance that will correspond to it for use in coordinating the distributed transaction. An XA connection instance is an instance of a class that implements the standard javax.sql.XAConnection interface:
public interface XAConnection extends PooledConnection { javax.jta.xa.XAResource getXAResource() throws SQLException; }

As you see, the XAConnection interface extends the javax.sql.PooledConnection interface, so it also includes the getConnection,

28-6 Oracle Database JDBC Developer's Guide and Reference

XA Components

close, addConnectionEventListener, and removeConnectionEventListener methods. Oracle JDBC implements the XAConnection interface with the OracleXAConnection class, located both in the oracle.jdbc.xa.client package and the oracle.jdbc.xa.server package. The OracleXAConnection classes also extend the OraclePooledConnection class. The OracleXAConnection class getXAResource method returns the Oracle implementation of an OracleXAResource instance, which is an OracleXAResource instance. The getConnection method returns an OracleConnection instance. A JDBC connection instance returned by an XA connection instance acts as a temporary handle to the physical connection, as opposed to encapsulating the physical connection. The physical connection is encapsulated by the XA connection instance. The connection obtained from an XAConnection object behaves exactly like a regular connection, until it participates in a global transaction. At that time, auto-commit status is set to false. After the global transaction ends, auto-commit status is returned to the value it had before the global transaction. The default auto-commit status on a connection obtained from XAConnection is false in all releases prior to Oracle Database 10g. In Oracle Database 10g, the default status is true. Each time an XA connection instance getConnection method is called, it returns a new connection instance that exhibits the default behavior, and closes any previous connection instance that still exists and had been returned by the same XA connection instance. However, it is advisable to explicitly close any previous connection instance before opening a new one. Calling the close method of an XA connection instance closes the physical connection to the database. This is typically performed in the middle tier.

XAResource Interface and Oracle Implementation

The transaction manager uses OracleXAResource instances to coordinate all the transaction branches that constitute a distributed transaction. Each OracleXAResource instance provides the following key functionality, typically invoked by the transaction manager:

It associates and disassociates distributed transactions with the transaction branch operating in the XA connection instance that produced this OracleXAResource instance. Essentially, it associates distributed transactions with the physical connection or session encapsulated by the XA connection instance. This is done through use of transaction IDs. It performs the two-phase COMMIT functionality of a distributed transaction to ensure that changes are not committed in one transaction branch before there is assurance that the changes will succeed in all transaction branches.

Distributed Transactions 28-7

XA Components

Notes:

Because there must always be a one-to-one correlation between XA connection instances and OracleXAResource instances, an OracleXAResource instance is implicitly closed when the associated XA connection instance is closed. If a transaction is opened by a given OracleXAResource instance, then it must also be closed by the same OracleXAResource instance.

An OracleXAResource instance is an instance of a class that implements the standard javax.transaction.xa.XAResource interface. Oracle JDBC implements the XAResource interface with the OracleXAResource class, located both in the oracle.jdbc.xa.client package and the oracle.jdbc.xa.server package. Oracle JDBC driver creates and returns an OracleXAResource instance whenever the getXAResource method of the OracleXAConnection class is called, and it is Oracle JDBC driver that associates an OracleXAResource instance with a connection instance and the transaction branch being run through that connection. This method is how an OracleXAResource instance is associated with a particular connection and with the transaction branch being run in that connection.

OracleXAResource Method Functionality and Input Parameters

The OracleXAResource class has several methods to coordinate a transaction branch with the distributed transaction with which it is associated. This functionality usually involves two-phase COMMIT operations. A transaction manager, receiving OracleXAResource instances from a middle-tier component, such as an application server, typically invokes this functionality. Each of these methods takes a transaction ID as input, in the form of an Xid instance, which includes a transaction branch ID component and a distributed transaction ID component. Every transaction branch has a unique transaction ID, but transaction branches belonging to the same global transaction have the same global transaction component as part of their transaction IDs. Start Starts work on behalf of a transaction branch, associating the transaction branch with a distributed transaction.
void start(Xid xid, int flags)

The flags parameter must be one or more of the following values:

XAResource.TMNOFLAGS Flag the start of a new transaction branch for subsequent operations in the session associated with this XA resource instance. This branch will have the transaction ID xid, which is an OracleXid instance created by the transaction manager. This will map the transaction branch to the appropriate distributed transaction.

XAResource.TMJOIN Join subsequent operations in the session associated with this XA resource instance to the existing transaction branch specified by xid.

28-8 Oracle Database JDBC Developer's Guide and Reference

XA Components

XAResource.TMRESUME Resume the transaction branch specified by xid.

Note:

A transaction branch can be resumed only if it had been suspended earlier.

OracleXAResource.ORATMSERIALIZABLE Start a serializable transaction with transaction ID xid.

OracleXAResource.ORATMREADONLY Start a read-only transaction with transaction ID xid.

OracleXAResource.ORATMREADWRITE Start a read/write transaction with transaction ID xid.

OracleXAResource.ORATRANSLOOSE Start a loosely coupled transaction with transaction ID xid.

TMNOFLAGS, TMJOIN, TMRESUME, ORATMSERIALIZABLE, ORATMREADONLY, and ORATMREADWRITE are defined as static members of the XAResource interface and OracleXAResource class. ORATMSERIALIZABLE, ORATMREADONLY, and ORATMREADWRITE are the isolation-mode flags. The default isolation behavior is READ COMMITTED.
Note:

Instead of using the start method with TMRESUME, the transaction manager can cast to OracleXAResource and use the resume(Xid xid) method, an Oracle extension. If you use TMRESUME, then you must also use TMNOMIGRATE, as in start(xid, XAResource.TMRESUME | OracleXAResource.TMNOMIGRATE). This prevents the application from receiving the error ORA 1002: fetch out of sequence. If you use the isolation-mode flags incorrectly, then an exception with code XAER_INVAL is raised. Furthermore, you cannot use isolation-mode flags when resuming a global transaction, because you cannot set the isolation level of an existing transaction. If you try to use the isolation-mode flags when resuming a transaction, then an external Oracle exception with code ORA-24790 is raised. In order to avoid Error ORA 1002: fetch out of sequence, include the TMNOMIGRATE flag as part of the start method. For example:
start(xid, XAResource.TMSUSPEND | OracleXAResource.TMNOMIGRATE);

All the flags defined in OracleXAResource are Oracle extensions. When writing a transaction manager that uses these flags, you should be mindful of this.

Distributed Transactions 28-9

XA Components

Note that to create an appropriate transaction ID in starting a transaction branch, the transaction manager must know which distributed transaction the transaction branch should belong to. The mechanics of this are handled between the middle tier and transaction manager. End Ends work on behalf of the transaction branch specified by xid, disassociating the transaction branch from its distributed transaction.
void end(Xid xid, int flags)

The flags parameter can have one of the following values:

XAResource.TMSUCCESS This is to indicate that this transaction branch is known to have succeeded.

XAResource.TMFAIL This is to indicate that this transaction branch is known to have failed.

XAResource.TMSUSPEND This is to suspend the transaction branch specified by xid. By suspending transaction branches, you can have multiple transaction branches in a single session. Only one can be active at any given time, however. Also, this tends to be more expensive in terms of resources than having two sessions.

TMSUCCESS, TMFAIL, and TMSUSPEND are defined as static members of the XAResource interface and OracleXAResource class.
Note:

Instead of using the end method with TMSUSPEND, the transaction manager can cast to OracleXAResource and use the suspend(Xid xid) method, an Oracle extension. This XA functionality to suspend a transaction provides a way to switch between various transactions within a single JDBC connection. You can use the XA classes to accomplish this, even if you are not in a distributed transaction environment and would otherwise have no need for the XA classes. If you use TMSUSPEND, then you must also use TMNOMIGRATE, as in end(xid, XAResource.TMSUSPEND | OracleXAResource.TMNOMIGRATE). This prevents the application from receiving the error ORA 1002: fetch out of sequence. In order to avoid Error ORA 1002: fetch out of sequence, include the TMNOMIGRATE flag as part of the end method. For example:
end(xid, XAResource.TMSUSPEND | OracleXAResource.TMNOMIGRATE);

All the flags defined in OracleXAResource are Oracle extensions. Any transaction manager that uses these flags should take heed of this.

28-10 Oracle Database JDBC Developer's Guide and Reference

XA Components

Prepare Prepares the changes performed in the transaction branch specified by xid. This is the first phase of a two-phase COMMIT operation, to ensure that the database is accessible and that the changes can be committed successfully.
int prepare(Xid xid)

This method returns an integer value as follows:

XAResource.XA_RDONLY This is returned if the transaction branch runs only read-only operations such as SELECT statements.

XAResource.XA_OK This is returned if the transaction branch runs updates that are all prepared without error.

NA (no value returned) No value is returned if the transaction branch runs updates and any of them encounter errors during preparation. In this case, an XA exception is thrown.

XA_RDONLY and XA_OK are defined as static members of the XAResource interface and OracleXAResource class.
Note:

Always call the end method on a branch before calling the prepare method. If there is only one transaction branch in a distributed transaction, then there is no need to call the prepare method. You can call the OracleXAResource commit method without preparing first.

Commit Commits prepared changes in the transaction branch specified by xid. This is the second phase of a two-phase COMMIT and is performed only after all transaction branches have been successfully prepared.
void commit(Xid xid, boolean onePhase)

Set the onePhase parameter as follows:

true This is to use one-phase instead of two-phase protocol in committing the transaction branch. This is appropriate if there is only one transaction branch in the distributed transaction; the prepare step would be skipped.

false This is to use two-phase protocol in committing the transaction branch.

Roll back Rolls back prepared changes in the transaction branch specified by xid.
void rollback(Xid xid)

Distributed Transactions

28-11

XA Components

Forget Tells the resource manager to forget about a heuristically completed transaction branch.
public void forget(Xid xid)

Recover The transaction manager calls this method during recovery to obtain the list of transaction branches that are currently in prepared or heuristically completed states.
public Xid[] recover(int flag)

Note: Values for flag other than TMSTARTRSCAN, TMENDRSCAN, or TMNOFLAGS, cause an exception to be thrown, otherwise flag is ignored.

The resource manager returns zero or more Xids for the transaction branches that are currently in a prepared or heuristically completed state. If an error occurs during the operation, then the resource manager throws the appropriate XAException.
Note: The recover method requires SELECT privilege on DBA_PENDING_TRANSACTIONS and EXECUTE privilege on SYS.DBMS_XA in Oracle database server. For database versions prior to Oracle Database 11g Release 1 (11.1), where an Oracle patch including fix for bug 5945463 is not available or it is infeasible to apply the patch for the particular application scenario, the recover method further requires EXECUTE privilege on SYS.DBMS_SYSTEM in Oracle database server.

Check for same RM To determine if two OracleXAResource instances correspond to the same resource manager, call the isSameRM method from one OracleXAResource instance, specifying the other OracleXAResource instance as input. In the following example, presume xares1 and xares2 are OracleXAResource instances:
boolean sameRM = xares1.isSameRM(xares2);

Xid Interface and Oracle Implementation

The transaction manager creates transaction ID instances and uses them in coordinating the branches of a distributed transaction. Each transaction branch is assigned a unique transaction ID, which includes the following information:

Format identifier A format identifier specifies a Java transaction manager. For example, there could be a format identifier ORCL. This field cannot be null. The size of a format identifier is 4 bytes.

Global transaction identifier It is also known as a distributed transaction ID component. The size of a global transaction identifier is 64 bytes.

Branch qualifier

28-12 Oracle Database JDBC Developer's Guide and Reference

Error Handling and Optimizations

It is also known as transaction branch ID component. The size of a branch qualifier is 64 bytes. The 64-byte global transaction identifier value will be identical in the transaction IDs of all transaction branches belonging to the same distributed transaction. However, the overall transaction ID is unique for every transaction branch. An XA transaction ID instance is an instance of a class that implements the standard javax.transaction.xa.Xid interface, which is a Java mapping of the X/Open transaction identifier XID structure. Oracle implements this interface with the OracleXid class in the oracle.jdbc.xa package. OracleXid instances are employed only in a transaction manager, transparent to application programs or an application server.
Note:

Oracle does not require the use of OracleXid for OracleXAResource calls. Instead, use any class that implements the javax.transaction.xa.Xid interface.

A transaction manager may use the following in creating an OracleXid instance:

public OracleXid(int fId, byte gId[], byte bId[]) throws XAException

Where fId is an integer value for the format identifier, gId[] is a byte array for the global transaction identifier, and bId[] is a byte array for the branch qualifier. The Xid interface specifies the following getter methods:

public int getFormatId() public byte[] getGlobalTransactionId() public type[] getBranchQualifier()

Error Handling and Optimizations

This section focuses on the functionality of XA exceptions and error handling and the Oracle optimizations in its XA implementation. It covers the following topics:

XAException Classes and Methods Mapping between Oracle Errors and XA Errors XA Error Handling Oracle XA Optimizations

The exception and error-handling discussion includes the standard XA exception class and the Oracle-specific XA exception class, as well as particular XA error codes and error-handling techniques.

XAException Classes and Methods

XA methods throw XA exceptions, as opposed to general exceptions or SQLExceptions. An XA exception is an instance of the standard class javax.transaction.xa.XAException or a subclass. Oracle subclasses XAException with the oracle.jdbc.xa.OracleXAException class. An OracleXAException instance consists of an Oracle error portion and an XA error portion and is constructed using one of the following constructors:

Distributed Transactions

28-13

Error Handling and Optimizations

public OracleXAException() public OracleXAException(int error)

The error value is an error code that combines an Oracle SQL error value and an XA error value. The JDBC driver determines exactly how to combine the Oracle and XA error values. The OracleXAException class has the following methods:

public int getOracleError() This method returns the Oracle SQL error code pertaining to the exception, a standard ORA error number or 0 if there is no Oracle SQL error.

public int getXAError() This method returns the XA error code pertaining to the exception. XA error values are defined in the javax.transaction.xa.XAException class.

Mapping between Oracle Errors and XA Errors

Oracle errors correspond to XA errors in OracleXAException instances as documented in Table 282.
Table 282 Oracle-XA Error Mapping XA Error Code XAException.XAER_RMFAIL XAException.XAER_RMFAIL XAException.XAER_NOTA XAException.XA_HEURCOM XAException.XA_HEURRB XAException.XA_HEURMIX XAException.XA_RDONLY XAException.XA_RETRY XAException.XAER_RMERR

Oracle Error Code ORA 3113 ORA 3114 ORA 24756 ORA 24764 ORA 24765 ORA 24766 ORA 24767 ORA 25351 all other ORA errors

XA Error Handling
The following example uses the OracleXAException class to process an XA exception:
try { ... ...Perform XA operations... ... } catch(OracleXAException oxae) { int oraerr = oxae.getOracleError(); System.out.println("Error " + oraerr); } catch(XAException xae) {...Process generic XA exception...}

In case the XA operations did not throw an Oracle-specific XA exception, the code drops through to process a generic XA exception.

28-14 Oracle Database JDBC Developer's Guide and Reference

Implementing a Distributed Transaction

Oracle XA Optimizations
Oracle JDBC has functionality to improve performance if two or more branches of a distributed transaction use the same database instance, meaning that the OracleXAResource instances associated with these branches are associated with the same resource manager. In such a circumstance, the prepare method of only one of these OracleXAResource instances will return XA_OK or will fail. The rest will return XA_RDONLY, even if updates are made. This allows the transaction manager to implicitly join all the transaction branches and commit or roll back, in case of failure, the joined transaction through the OracleXAResource instance that returned XA_OK or failed. The transaction manager can use the OracleXAResource class isSameRM method to determine if two OracleXAResource instances are using the same resource manager. This way it can interpret the meaning of XA_RDONLY return values.

Implementing a Distributed Transaction

This section provides an example of how to implement a distributed transaction using Oracle XA functionality. This section covers the following topics:

Summary of Imports for Oracle XA Oracle XA Code Sample

Summary of Imports for Oracle XA

You must import the following for Oracle XA functionality:
import import import import import oracle.jdbc.xa.OracleXid; oracle.jdbc.xa.OracleXAException; oracle.jdbc.pool.*; oracle.jdbc.xa.client.*; javax.transaction.xa.*;

The oracle.jdbc.pool package has classes for connection pooling functionality, some of which are subclassed by XA-related classes. Alternatively, if the code will run inside Oracle Database and access that database for SQL operations, you must import oracle.jdbc.xa.server instead of oracle.jdbc.xa.client.
import oracle.jdbc.xa.server.*;

If your application must access another Oracle Database as part of an XA transaction using the server-side Thin driver, then your code can use the fully qualified names of the oracle.xa.client classes. The client and server packages each have versions of the OracleXADataSource, OracleXAConnection, and OracleXAResource classes. Abstract versions of these three classes are in the top-level oracle.jdbc.xa package.

Oracle XA Code Sample

This example uses a two-phase distributed transaction with two transaction branches, each to a separate database.

Distributed Transactions

28-15

Implementing a Distributed Transaction

Note that for simplicity, this example combines code that would typically be in a middle tier with code that would typically be in a transaction manager, such as the OracleXAResource method invocations and the creation of transaction IDs. For brevity, the specifics of creating transaction IDs and performing SQL operations are not shown here. The complete example is shipped with the product. This example performs the following sequence:
1. 2. 3. 4. 5. 6. 7. 8. 9.

Start transaction branch #1. Start transaction branch #2. Execute DML operations on branch #1. Execute DML operations on branch #2. End transaction branch #1. End transaction branch #2. Prepare branch #1. Prepare branch #2. Commit branch #1.

10. Commit branch #2. // You import import import import import import import import need to import the java.sql package to use JDBC java.sql.*; javax.sql.*; oracle.jdbc.*; oracle.jdbc.pool.*; oracle.jdbc.xa.OracleXid; oracle.jdbc.xa.OracleXAException; oracle.jdbc.xa.client.*; javax.transaction.xa.*;

class XA4 { public static void main (String args []) throws SQLException { try { String URL1 = "jdbc:oracle:oci:@"; // You can put a database name after the @ sign in the connection URL. String URL2 ="jdbc:oracle:thin:@(description=(address=(host=dlsun991) (protocol=tcp)(port=5521))(connect_data=(sid=rdbms2)))"; // Create first DataSource and get connection OracleDataSource ods1 = new OracleDataSource(); ods1.setURL(URL1); ods1.setUser("scott"); ods1.setPassword("tiger"); Connection conna = ods1.getConnection(); // Create second DataSource and get connection OracleDataSource ods2 = new OracleDataSource(); ods2.setURL(URL2); ods2.setUser("scott"); ods2.setPassword("tiger"); Connection connb = ods2.getConnection();

28-16 Oracle Database JDBC Developer's Guide and Reference

Implementing a Distributed Transaction

// Prepare a statement to create the table Statement stmta = conna.createStatement (); // Prepare a statement to create the table Statement stmtb = connb.createStatement (); try { // Drop the test table stmta.execute ("drop table my_table"); } catch (SQLException e) { // Ignore an error here } try { // Create a test table stmta.execute ("create table my_table (col1 int)"); } catch (SQLException e) { // Ignore an error here too } try { // Drop the test table stmtb.execute ("drop table my_tab"); } catch (SQLException e) { // Ignore an error here } try { // Create a test table stmtb.execute ("create table my_tab (col1 char(30))"); } catch (SQLException e) { // Ignore an error here too } // Create XADataSource instances and set properties. OracleXADataSource oxds1 = new OracleXADataSource(); oxds1.setURL("jdbc:oracle:oci:@"); oxds1.setUser("scott"); oxds1.setPassword("tiger"); OracleXADataSource oxds2 = new OracleXADataSource(); oxds2.setURL("jdbc:oracle:thin:@(description=(address=(host=dlsun991) (protocol=tcp)(port=5521))(connect_data=(sid=rdbms2)))"); oxds2.setUser("scott"); oxds2.setPassword("tiger"); // Get XA connections to the underlying data sources

Distributed Transactions

28-17

Implementing a Distributed Transaction

XAConnection pc1 XAConnection pc2

= oxds1.getXAConnection(); = oxds2.getXAConnection();

// Get the physical connections Connection conn1 = pc1.getConnection(); Connection conn2 = pc2.getConnection(); // Get the XA resources XAResource oxar1 = pc1.getXAResource(); XAResource oxar2 = pc2.getXAResource(); // Create the Xids With the Same Global Ids Xid xid1 = createXid(1); Xid xid2 = createXid(2); // Start the Resources oxar1.start (xid1, XAResource.TMNOFLAGS); oxar2.start (xid2, XAResource.TMNOFLAGS); // Execute SQL operations with conn1 and conn2 doSomeWork1 (conn1); doSomeWork2 (conn2); // END both the branches -- IMPORTANT oxar1.end(xid1, XAResource.TMSUCCESS); oxar2.end(xid2, XAResource.TMSUCCESS); // Prepare the RMs int prp1 = oxar1.prepare (xid1); int prp2 = oxar2.prepare (xid2); System.out.println("Return value of prepare 1 is " + prp1); System.out.println("Return value of prepare 2 is " + prp2); boolean do_commit = true; if (!((prp1 == XAResource.XA_OK) || (prp1 == XAResource.XA_RDONLY))) do_commit = false; if (!((prp2 == XAResource.XA_OK) || (prp2 == XAResource.XA_RDONLY))) do_commit = false; System.out.println("do_commit is " + do_commit); System.out.println("Is oxar1 same as oxar2 ? " + oxar1.isSameRM(oxar2)); if (prp1 == XAResource.XA_OK) if (do_commit) oxar1.commit (xid1, false); else oxar1.rollback (xid1); if (prp2 == XAResource.XA_OK) if (do_commit) oxar2.commit (xid2, false); else oxar2.rollback (xid2); // Close connections conn1.close(); conn1 = null;

28-18 Oracle Database JDBC Developer's Guide and Reference

Implementing a Distributed Transaction

conn2.close(); conn2 = null; pc1.close(); pc1 = null; pc2.close(); pc2 = null; ResultSet rset = stmta.executeQuery ("select col1 from my_table"); while (rset.next()) System.out.println("Col1 is " + rset.getInt(1)); rset.close(); rset = null; rset = stmtb.executeQuery ("select col1 from my_tab"); while (rset.next()) System.out.println("Col1 is " + rset.getString(1)); rset.close(); rset = null; stmta.close(); stmta = null; stmtb.close(); stmtb = null; conna.close(); conna = null; connb.close(); connb = null; } catch (SQLException sqe) { sqe.printStackTrace(); } catch (XAException xae) { if (xae instanceof OracleXAException) { System.out.println("XA Error is " + ((OracleXAException)xae).getXAError()); System.out.println("SQL Error is " + ((OracleXAException)xae).getOracleError()); } } } static Xid createXid(int bids) throws XAException {...Create transaction IDs...} private static void doSomeWork1 (Connection conn) throws SQLException {...Execute SQL operations...} private static void doSomeWork2 (Connection conn) throws SQLException {...Execute SQL operations...} }

Distributed Transactions

28-19

Native-XA in Oracle JDBC Drivers

Native-XA in Oracle JDBC Drivers

In general, XA commands can be sent to the server in the following ways:

Through PL/SQL procedures. However, this method is very slow. Through native APIs. This method is faster than using PL/SQL procedures.

There is a huge performance difference between the two methods of sending XA commands to the server. The use of native APIs provides high performance gains as compared to the use of PL/SQL procedures. Prior to Oracle Database 10g, the Thin driver used PL/SQL procedures to send XA commands to the server because Thin native APIs were not available. In Oracle Database 10g, the Thin native APIs are available and are used by default to send XA commands. This section covers the following topics:

OCI Native XA Thin Native XA

OCI Native XA
Native XA is enabled through the use of the tnsEntry and nativeXA properties of the OracleXADataSource class.
See Also: Table 82, " Oracle Extended Data Source Properties" on page 8-4 for explanation of these properties.

Note:

Currently, OCI Native XA does not work in a multithreaded environment. The OCI driver uses the C/XA library of Oracle to support distributed transactions, which requires that an XAConnection be obtained for each thread before resuming a global transaction.

Configuration and Installation On a Sun Solaris or Linux system, you need the shared libraries, libheteroxa11.so and libheteroxa11_g.so, to enable the Native XA feature. In order for the Native XA feature to work properly, these libraries need to be installed and available in the Sun Solaris search path. On a Microsoft Windows system, you need the heteroxa11.dll and heteroxa11_g.dll files to enable the Native XA feature. These files need to be installed and available in the Windows DLL path for the Native XA feature to work properly.
Note:

Libraries with the _g suffix are debug libraries.

Exception Handling When using the Native XA feature in distributed transactions, it is recommended that the application simply check for XAException or SQLException, rather than OracleXAException or OracleSQLException.

28-20 Oracle Database JDBC Developer's Guide and Reference

Native-XA in Oracle JDBC Drivers

See Also: "Native XA Messages" on page C-9 for a listing of Native XA messages.

Note:

The mapping from SQL error codes to standard XA error codes does not apply to the Native XA feature.

Native XA Code Example The following portion of code shows how to enable the Native XA feature:
... // Create a XADataSource instance OracleXADataSource oxds = new OracleXADataSource(); oxds.setURL(url); // Set the nativeXA property to use Native XA feature oxds.setNativeXA(true); // Set the tnsEntry property to an older DB as required oxds.setTNSEntryName("ora805"); ...

Thin Native XA
Like the JDBC OCI driver, the JDBC Thin driver also provides support for Native XA. However, the JDBC Thin driver provides support for Native XA by default. This is unlike the case of the JDBC OCI driver in which the support for Native XA is not enabled by default. You can disable Native XA by calling setNativeXA(false) on the XA data source as follows:
... // Create a XADataSource instance OracleXADataSource oxds = new OracleXADataSource(); ... // Disabling Native XA oxds.setNativeXA(false); ...

For example, you may need to disable Native XA as a workaround for a bug in the Native XA code.

Distributed Transactions

28-21

Native-XA in Oracle JDBC Drivers

28-22 Oracle Database JDBC Developer's Guide and Reference

Part VIII
Manageability
This part discusses the database management and diagnosability support in Oracle Java Database Connectivity (JDBC) drivers. Part VIII contains the following chapters:

Chapter 29, "Database Management" Chapter 30, "Diagnosability in JDBC" Chapter 31, "JDBC DMS Metrics"

29
Database Management
Oracle Database 11g Release 1 (11.1) provides many new features for managing the database. This chapter describes the following:

Database Startup and Shutdown Database Change Notification

Database Startup and Shutdown

Oracle Database 11g Release 1 (11.1) introduces two new JDBC methods, startup and shutdown, in oracle.jdbc.OracleConnection that enable you to start up and shut down an Oracle Database instance. This is similar to the way you would start up or shut down a database instance from SQL*Plus. To use these methods, you must have a dedicated connection to the server. You cannot be connected to a shared server through a dispatcher. To use the startup and shutdown methods, you must be connected as SYSDBA or SYSOPER. To connect as SYSDBA or SYSOPER with Oracle JDBC drivers, you need to set the internal_logon connection property accordingly. To logon as SYSDBA with the JDBC Thin driver you need to configure the server to use the password file. For example, to configure system/manager to connect as sysdba with the JDBC Thin driver, perform the following:
1.

From the command line, type:

orapwd file=$ORACLE_HOME/dbs/orapw entries=5 Enter password: password

2.

Connect to database as SYSDBA and run the following commands from SQL*Plus:
GRANT SYSDBA TO system; PASSWORD system Changing password for system New password: password Retype new password: password

3.

Edit init.ora and add the following line:

REMOTE_LOGIN_PASSWORDFILE=EXCLUSIVE

4.

Restart the database instance.

As opposed to the JDBC Thin driver, the JDBC OCI driver can connect as SYSDBA or SYSOPER locally without specifying a password file on the server.

Database Management

29-1

Database Startup and Shutdown

To start a database instance using the startup method, the application must first connect to the database as a SYSDBA or SYSOPER in the PRELIM_AUTH mode, which is the only connection mode that is permitted when the database is down. You can do this by setting the new connection property prelim_auth to true. In the PRELIM_ AUTH mode, you can only start up a database instance that is down. You cannot run any SQL statements in this mode. The startup method only starts up a database instance. It does not mount it nor open it. You have to reconnect as SYSDBA or SYSOPER, but without the PRELIM_AUTH mode, and run the following commands to mount and open the database instance:
ALTER DATABASE MOUNT ALTER DATABASE OPEN

Note: The startup method will start up the database using the server parameter file. Oracle JDBC drivers do not support database startup using client parameter file.

The startup method takes a parameter that specifies the database startup option. Table 291 lists the supported database startup options. These options are defined in the oracle.jdbc.OracleConnection.DatabaseStartupMode class.
Table 291 Option FORCE NO_RESTRICTION RESTRICT Supported Database Startup Options Description Shuts down the database in the abort mode before starting a new instance. Starts up the database with no restrictions. Starts up the database and allows database access only to users with both the CREATE SESSION and RESTRICTED SESSION privileges (usually, the DBA).

The shutdown method enables you to shut down an Oracle Database instance. To use this method, you must be connected to the database as a SYSDBA or SYSOPER. Like the startup method, the shutdown method also takes a parameter. In this case, the parameter specifies the database shutdown option. Table 292 lists the supported database shutdown options. These options are defined in the oracle.jdbc.OracleConnection.DatabaseShutdownMode class.
Table 292 Option ABORT CONNECT FINAL IMMEDIATE TRANSACTIONAL TRANSACTIONAL_LOCAL Supported Database Shutdown Options Description Does not wait for current calls to complete or users to disconnect from the database. Refuses any new connection and waits for existing connection to end. Shuts down the database. Does not wait for current calls to complete or users to disconnect from the database. Refuses new transactions and waits for active transactions to end. Refuses new local transactions and waits for active local transactions to end.

For shut down options other than ABORT and FINAL, you must call the shutdown method again with the FINAL option to actually shut down the database.

29-2 Oracle Database JDBC Developer's Guide and Reference

Database Startup and Shutdown

Note: The shutdown(DatabaseShutdownMode.FINAL) method must be preceded by another call to the shutdown method with one of the following options: CONNECT, TRANSACTIONAL, TRANSACTIONAL_LOCAL, or IMMEDIATE. Otherwise the call hangs.

A standard way to shut down the database is as follows:

1.

Initiate shutdown by prohibiting further connections or transactions in the database. The shut down option can be either CONNECT, TRANSACTIONAL, TRANSACTIONAL_LOCAL, or IMMEDIATE. Dismount and close the database by calling the appropriate ALTER DATABASE command. Finish shutdown using the FINAL option.

2. 3.

In special circumstances to shutdown the database as fast as possible, the ABORT option can be used. This is the equivalent to SHUTDOWN ABORT in SQL*Plus. Example Example 291 illustrates the use of the startup and shutdown methods.
Example 291 Database Startup and Shutdown

import java.sql.Statement; import java.util.Properties; import oracle.jdbc.OracleConnection; import oracle.jdbc.pool.OracleDataSource; /** * To logon as sysdba, you need to create a password file for user "sys": * orapwd file=/path/orapw password=password entries=300 * and add the following setting in init.ora: * REMOTE_LOGIN_PASSWORDFILE=EXCLUSIVE * then restart the database. */ public class DBStartup { static final String DB_URL = "jdbc:oracle:thin:@(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=XYZ.com)(PORT=1521))" + "(CONNECT_DATA=(SERVICE_NAME=rdbms.devplmt.XYZ.com)))"; public static void main(String[] argv) throws Exception { // Starting up the database: OracleDataSource ds = new OracleDataSource(); Properties prop = new Properties(); prop.setProperty("user","sys"); prop.setProperty("password","manager"); prop.setProperty("internal_logon","sysdba"); prop.setProperty("prelim_auth","true"); ds.setConnectionProperties(prop); ds.setURL(DB_URL); OracleConnection conn = (OracleConnection)ds.getConnection(); conn.startup(OracleConnection.DatabaseStartupMode.NO_RESTRICTION); conn.close(); // Mounting and opening the database OracleDataSource ds1 = new OracleDataSource(); Properties prop1 = new Properties(); Database Management 29-3

Database Change Notification

prop1.setProperty("user","sys"); prop1.setProperty("password","manager"); prop1.setProperty("internal_logon","sysdba"); ds1.setConnectionProperties(prop1); ds1.setURL(DB_URL); OracleConnection conn1 = (OracleConnection)ds1.getConnection(); Statement stmt = conn1.createStatement(); stmt.executeUpdate("ALTER DATABASE MOUNT"); stmt.executeUpdate("ALTER DATABASE OPEN"); stmt.close(); conn1.close(); // Shutting down the database OracleDataSource ds2 = new OracleDataSource(); Properties prop = new Properties(); prop.setProperty("user","sys"); prop.setProperty("password","manager"); prop.setProperty("internal_logon","sysdba"); ds2.setConnectionProperties(prop); ds2.setURL(DB_URL); OracleConnection conn2 = (OracleConnection)ds2.getConnection(); conn2.shutdown(OracleConnection.DatabaseShutdownMode.IMMEDIATE); Statement stmt1 = conn2.createStatement(); stmt1.executeUpdate("ALTER DATABASE CLOSE NORMAL"); stmt1.executeUpdate("ALTER DATABASE DISMOUNT"); stmt1.close(); conn2.shutdown(OracleConnection.DatabaseShutdownMode.FINAL); conn2.close(); } }

Database Change Notification

Generally, a mid-tier data cache duplicates some data from the back-end database server. Its goal is to avoid redundant queries to the database. However, this is efficient only when the data rarely changes in the database. The data cache has to be updated or invalidated when the data changes in the database. The 11g release 1 (11.1) Oracle JDBC drivers provide support for the Database Change Notification feature of Oracle Database. Using this functionality of the JDBC drivers, multi-tier systems can take advantage of the Database Change Notification feature to maintain a data cache as updated as possible by receiving invalidation events from the JDBC drivers. The JDBC drivers can register SQL queries with the database and receive notifications in response to:

DML or DDL changes on the objects associated with the queries. DML or DDL changes that affect the result set.

The notifications are published when the DML or DDL transaction commits (changes made in a local transaction do not generate any event until they are comitted). To use Oracle JDBC driver support for Database Change Notification, you need to perform the following:
1. 2.

Registration: You first need to create a registration Query association: After you have created a registration, you can associate SQL queries with it. These queries are part of the registration.

29-4 Oracle Database JDBC Developer's Guide and Reference

Database Change Notification

3.

Notification: Notifications are created in response to changes in tables or result set. Oracle database communicates these notifications to the JDBC drivers through a dedicated network connection and JDBC drivers convert these notifications to Java events.

Also, you need to grant the CHANGE NOTIFICATION privilege to the user. For example, if you connect to the database using the SCOTT user name, then you need to run the following command in the database:
grant change notification to scott;

Creating a Registration Creating a registration is a one-time process and is done outside of the currently used transaction. The API for creating a registration in the server is executed in its own transaction and is committed immediately. You need a JDBC connection to create a registration, however, the registration is not attached to the connection. You can close the connection after creating a registration, and the registration survives. In a RAC environment, a registration is a persistent entity that exists on all nodes. If a node goes down, then the registration continues to exist and will be notified when the tables change. There are two ways to create a registration:

The JDBC-style of registration: Use the JDBC driver to create a registration on the server. The JDBC driver launches a new thread that listens to notifications from the server (through a dedicated channel) and converts these notification messages into Java events. The driver then notifies all the listeners registered with this registration. The PLSQL-style of registration: If you want a PLSQL stored procedure to handle the notifications, then create a PLSQL-style registration. Like in the JDBC-style of registration, the JDBC drivers enable you to attach statements (queries) to this registration. However the JDBC drivers do not get notifications from the server because the notifications are handled by the PLSQL stored procedure.
Note:

There is no way to remove one particular object (table) from an existing registration. A workaround would be to either create a new registration without this object or ignore the events that are related to this object.

You can use the registerDatabaseChangeNotification method of the oracle.jdbc.OracleConnection interface to create a JDBC-style of registration. You can set certain registration options through the options parameter of this method. Table 293 lists some of the registration options that can be set. To set these options, use the java.util.Properties object. These options are defined in the oracle.jdbc.OracleConnection interface. The registration options have a direct impact on the notification events that the JDBC drivers will create. Example 291 illustrates how to use the Database Change Notification feature. The registerDatabaseChangeNotification method creates a new database change registration in the database server with the given options. It returns a DatabaseChangeRegistration object, which can then be used to associate a statement with this registration. It also opens a listener socket that will be used by the database to send notifications.

Database Management

29-5

Database Change Notification

Note:

If a listener socket (created by a different registration) exists, then this socket will be used by the new database change registration as well.

Table 293 Option

Database Change Notification Registration Options Description If set to true, DELETE operations will not generate any database change event. If set to true, INSERT operations will not generate any database change event. If set to true, UPDATE operations will not generate any database change event. Specifies the number of transactions by which the client is willing to lag behind. Note: If this option is set to any value other than 0, then ROWID level granularity of information will not be available in the events, even if the DCN_NOTIFY_ROWIDS option is set to true.

DCN_IGNORE_DELETEOP DCN_IGNORE_INSERTOP DCN_IGNORE_UPDATEOP DCN_NOTIFY_CHANGELAG

DCN_NOTIFY_ROWIDS

Database change events will include row-level details, such as operation type and ROWID.

DCN_QUERY_CHANGE_NOTIFICATION Activates query change notification instead of object change notification. Note: This option is only available when running against an 11.0 database. NTF_LOCAL_HOST NTF_LOCAL_TCP_PORT NTF_QOS_PURGE_ON_NTFN NTF_QOS_RELIABLE NTF_TIMEOUT Specifies the IP address of the computer that will receive the notifications from the server. Specifies the TCP port that the driver should use for the listener socket. Specifies whether the registration should be expunged on the first notification event. Specifies whether to make the notifications persistent, which comes at a performance cost. Specifies the time in seconds after which the registration will be automatically expunged by the database.

If there exists an registration, then you can also use the getDatabaseChangeRegistration method to map the existing registration with a new DatabaseChangeRegistration object. This method is particularly useful if you have created a registration using PL/SQL and want to associate a statement with it.
See:

Refer to the JavaDoc for more information about the APIs.

Associating a Query With a Registration After you have created a registration or mapped to an existing registration, you can associate a query with it. Like creating a registration, associating a query with a registration is a one-time process and is done outside of the currently used registration. The query will be associated even if the local transaction is rolled back. You can associate a query with registration using the setDatabaseChangeRegistration method defined in the OracleStatement class. This method takes a DatabaseChangeRegistration object as parameter. The following code snippet illustrates how to associate a query with a registration:

29-6 Oracle Database JDBC Developer's Guide and Reference

Database Change Notification

... // conn is a OracleConnection object. // prop is a Properties object containing the registration options. DatabaseChangeRegistration dcr = conn.registerDatabaseChangeNotifictaion(prop); ... Statement stmt = conn.createStatement(); // associating the query with the registration ((OracleStatement)stmt).setDatabaseChangeRegistration(dcr); // any query that will be executed with the 'stmt' object will be associated with // the registration 'dcr' until 'stmt' is closed or // '((OracleStatement)stmt).setDatabaseChangeRegistration(null);' is executed. ...

Notifying Database Change Events To receive database change notifications, you need to attach a listener to the registration. When a database change event occurs, the database server notifies the JDBC driver. The driver then constructs a new java event, identifies the registration to be notified, and notifies the listeners attached to the registration. The event contains the object ID of the database object that has changed and the type of operation that caused the change. Depending on the registration options, the event may also contain row-level detail information. The listener code can then use the event to make decisions about the data cache.
Note: The listener code must not slow down the JDBC notification mechanism. If the code is time-consuming, for example, if it refreshes the data cache by querying the database, then it needs to be executed within its own thread.

You can attach a listener to a registration using the addListener method. The following code snippet illustrates how to attach a listener to a registration:
... // conn is a OracleConnection object. // prop is a Properties object containing the registration options. DatabaseChangeRegistration dcr = conn.registerDatabaseChangeNotifictaion(prop); ... // Attach the listener to the registration. // Note: DCNListener is a custom listener and not a predefined or standard // lsiener DCNListener list = new DCNListener(); dcr.addListener(list); ...

Deleting a Registration You need to explicitly unregister a registration to delete it from the server and release the resources in the driver. You can unregister a registration using a connection different from one that was used for creating it. To unregister a registration, you can use the unregisterDatabaseChangeNotification method defined in oracle.jdbc.OracleConnection. You must pass the DatabaseChangeRegistration object as a parameter to this method. This method deletes the registration from the server and the driver and closes the listener socket.

Database Management

29-7

Database Change Notification

If the registration was created outside of JDBC, say using PL/SQL, then you need to pass the registration ID instead of the DatabaseChangeRegistration object. The method will delete the registration from the server, however, it does not free any resources in the driver. Example Example 292 illustrates how to use the Database Change Notification feature. In this example, the SCOTT user is connecting to the database. Therefore in the database you need to grant the following privilege to the user:
grant change notification to scott;

Example 292 import import import import import import import import import import

Database Change Notification

java.sql.ResultSet; java.sql.SQLException; java.sql.Statement; java.util.Properties; oracle.jdbc.OracleConnection; oracle.jdbc.OracleDriver; oracle.jdbc.OracleStatement; oracle.jdbc.dcn.DatabaseChangeEvent; oracle.jdbc.dcn.DatabaseChangeListener; oracle.jdbc.dcn.DatabaseChangeRegistration;

public class DBChangeNotification { static final String USERNAME= "scott"; static final String PASSWORD= "tiger"; static String URL; public static void main(String[] argv) { if(argv.length < 1) { System.out.println("Error: You need to provide the URL in the first argument."); System.out.println(" For example: > java -classpath .:ojdbc5.jar DBChangeNotification \"jdbc:oracle:thin: @(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=yourhost.yourdomain.com)(PORT=1521))(CO NNECT_DATA= (SERVICE_NAME=yourservicename)))\""); System.exit(1); } URL = argv[0]; DBChangeNotification demo = new DBChangeNotification(); try { demo.run(); } catch(SQLException mainSQLException ) { mainSQLException.printStackTrace(); } } void run() throws SQLException {

29-8 Oracle Database JDBC Developer's Guide and Reference

Database Change Notification

OracleConnection conn = connect(); // first step: create a registration on the server: Properties prop = new Properties(); // if connected through the VPN, you need to provide the TCP address of the client. // For example: // prop.setProperty(OracleConnection.NTF_LOCAL_HOST,"14.14.13.12"); // Ask the server to send the ROWIDs as part of the DCN events (small performance // cost): prop.setProperty(OracleConnection.DCN_NOTIFY_ROWIDS,"true"); // The following operation does a roundtrip to the database to create a new // registration for DCN. It sends the client address (ip address and port) that // the server will use to connect to the client and send the notification // when necessary. Note that for now the registration is empty (we haven't registered // any table). This also opens a new thread in the drivers. This thread will be // dedicated to DCN (accept connection to the server and dispatch the events to // the listeners). DatabaseChangeRegistration dcr = conn.registerDatabaseChangeNotification(prop); try { // add the listenerr: DCNDemoListener list = new DCNDemoListener(this); dcr.addListener(list); // second step: add objects in the registration: Statement stmt = conn.createStatement(); // associate the statement with the registration: ((OracleStatement)stmt).setDatabaseChangeRegistration(dcr); ResultSet rs = stmt.executeQuery("select * from dept where deptno='45'"); while (rs.next()) {} String[] tableNames = dcr.getTables(); for(int i=0;i<tableNames.length;i++) System.out.println(tableNames[i]+" is part of the registration."); rs.close(); stmt.close(); } catch(SQLException ex) { // if an exception occurs, we need to close the registration in order // to interrupt the thread otherwise it will be hanging around. if(conn != null) conn.unregisterDatabaseChangeNotification(dcr); throw ex; } finally { try {

Database Management

29-9

Database Change Notification

// Note that we close the connection! conn.close(); } catch(Exception innerex){ innerex.printStackTrace(); } } synchronized( this ) { // The following code modifies the dept table and commits: try { OracleConnection conn2 = connect(); conn2.setAutoCommit(false); Statement stmt2 = conn2.createStatement(); stmt2.executeUpdate("insert into dept (deptno,dname) values ('45','cool dept')",Statement.RETURN_GENERATED_KEYS); ResultSet autoGeneratedKey = stmt2.getGeneratedKeys(); if(autoGeneratedKey.next()) System.out.println("inserted one row with ROWID="+autoGeneratedKey.getString(1)); stmt2.executeUpdate("insert into dept (deptno,dname) values ('50','fun dept')",Statement.RETURN_GENERATED_KEYS); autoGeneratedKey = stmt2.getGeneratedKeys(); if(autoGeneratedKey.next()) System.out.println("inserted one row with ROWID="+autoGeneratedKey.getString(1)); stmt2.close(); conn2.commit(); conn2.close(); } catch(SQLException ex) { ex.printStackTrace(); } // wait until we get the event try{ this.wait();} catch( InterruptedException ie ) {} } // At the end: close the registration (comment out these 3 lines in order // to leave the registration open). OracleConnection conn3 = connect(); conn3.unregisterDatabaseChangeNotification(dcr); conn3.close(); } /** * Creates a connection the database. */ OracleConnection connect() throws SQLException { OracleDriver dr = new OracleDriver(); Properties prop = new Properties(); prop.setProperty("user",DBChangeNotification.USERNAME); prop.setProperty("password",DBChangeNotification.PASSWORD); return (OracleConnection)dr.connect(DBChangeNotification.URL,prop); } } /** * DCN listener: it prints out the event details in stdout. */ class DCNDemoListener implements DatabaseChangeListener {

29-10 Oracle Database JDBC Developer's Guide and Reference

Database Change Notification

DBChangeNotification demo; DCNDemoListener(DBChangeNotification dem) { demo = dem; } public void onDatabaseChangeNotification(DatabaseChangeEvent e) { Thread t = Thread.currentThread(); System.out.println("DCNDemoListener: got an event ("+this+" running on thread "+t+")"); System.out.println(e.toString()); synchronized( demo ){ demo.notify();} } }

This code will also work with Oracle Database 10g release 2 (10.2). This code uses "table registration". That is, when you register a SELECT query, what you register is the name of the tables involved and not the query itself. In other words you might select one single row of a table and if another row is updated, you will be notified although the result of your query has not changed. When using Oracle Database 11g, you can use a different option, the "query registration" with finer granularity. This can be done by setting the DCN_QUERY_ CHANGE_NOTIFICATION option. In this example, if you leave the registration open instead of closing it, then the database change notification thread continues to run. Now if you run a DML query that changes the SCOTT.DEPT table and commit it, say from SQL*Plus, then the java program prints the notification.

Database Management 29-11

Database Change Notification

29-12 Oracle Database JDBC Developer's Guide and Reference

30
Diagnosability in JDBC
In Oracle Database 11g, the JDBC drivers have been enhanced by including new diagnosabilty features and improving existing diagnosabilty features. These features enable users to diagnose problems in the applications that use Oracle JDBC drivers and the problems in the drivers themselves. They also reduce the effort required to develop and maintain Java applications that access an Oracle Database instance using Oracle JDBC drivers. Oracle JDBC drivers provide the following diagnosabilty features that enable users to identify and fix problems in their JDBC applications:

Logging Diagnosability Management The diagnosability features of the JDBC drivers are based on the standard java.util.logging framework and the javax.management MBean framework. Information about these standard frameworks is not covered in this document. Readers are advised to refer to the Sun Microsystems Web site to obtain information about these standard frameworks. Also refer to the JavaDoc for java.util.logging for information about the various configuration options available.
Note:

Logging
This feature logs information about events that occur when JDBC driver code runs. Events can include user-visible events, such as SQLExceptions, running of SQL statements, and detailed JDBC internal events, such as entry to and exit from internal JDBC methods. User can enable this feature to log specific events or all the events. Prior to Oracle Database 11g, JDBC drivers supported J2SE 2.0 and 3.0. These versions of J2SE did not include java.util.logging. Therefore, the logging feature provided by JDBC driver releases prior to Oracle Database 11g, differs from the java.util.logging framework. In Oracle Database 11g, the JDBC drivers no longer support J2SE 2.0 and 3.0. Therefore, the logging feature of JDBC drivers makes full use of the standard java.util.logging package. The enhanced logging system makes effective use of log levels to enable users to restrict log output to things of interest. It logs specific classes of information more consistently, making it easier for the user to understand the log file. This feature does not introduce new APIs or configuration files. Only new parameters are added to the existing standard java.util.logging configuration file. These
Diagnosability in JDBC 30-1

Logging

parameters are identical in use to the existing parameters and are intrinsic to using java.util.logging.
Note:

Oracle does not guarantee the exact content of the generated logs. To a large extent the log content is dependent on the details of the implementation. The details of the implementation change with every release, and therefore, the exact content of the logs are likely to change from release to release.

Enabling and Using JDBC Logging

Before you can start debugging your Java application, you need to enable and configure JDBC logging. This section covers the steps you need to perform to enable and use JDBC logging. It describes the following:

Configuring the CLASSPATH Environment Variable Enabling Logging Configuring Logging Using Loggers An Example

Configuring the CLASSPATH Environment Variable

Oracle ships several JAR files for each version of the JDBC drivers. The optimized JAR files do not contain any logging code and, therefore, do not generate any log output when used. To get log output, you must use the debug JAR files, which are indicated with a "_g" in the file name, like ojdbc5_g.jar or ojdbc6_g.jar. The debug JAR file must be included in the CLASSPATH environment variable.
Note: Ensure that the debug JAR file, say ojdbc5_g.jar or ojdbc6_g.jar, is the only Oracle JDBC JAR file in the CLASSPATH environment variable.

Enabling Logging
You can enable logging in the following ways:

Setting a Java system property You can enable logging by setting the oracle.jdbc.Trace system property.
java -Doracle.jdbc.Trace=true ...

Setting the system property enables global logging, which means that logging is enabled for the entire application. You can use global logging if you want to debug the entire application, or if you cannot or do not want to change the source code of the application.

Programmatically You can programmatically enable or disable logging in the following way: First, get the ObjectName of the Diagnosability MBean. The ObjectName is
com.oracle.jdbc:type=diagnosability,name=<loader>

30-2 Oracle Database JDBC Developer's Guide and Reference

Logging

where, loader is a filtered version of the result of calling toString on the context ClassLoader. Now, write the following lines of code:
// compute the ObjectName String loader = Thread.currentThread().getContextClassLoader().toString().replaceAll("[,=:\"]+" , ""); javax.management.ObjectName name = new javax.management.ObjectName("com.oracle.jdbc:type=diagnosability,name="+loader) ; // get the MBean server javax.management.MBeanServer mbs = java.lang.management.ManagementFactory.getPlatformMBeanServer(); // find out if logging is enabled or not System.out.println("LoggingEnabled = " + mbs.getAttribute(name, "LoggingEnabled")); // enable logging mbs.setAttribute(name, new javax.management.Attribute("LoggingEnabled", true)); // disable logging mbs.setAttribute(name, new javax.management.Attribute("LoggingEnabled", false));

Programmatic enabling and disabling of logging helps you to control what parts of your application need to generate log output.
Note:

Enabling logging using either of the methods would only generate a minimal log of serious errors. Usually this is not of much use. To generate and more useful and detail log, you must configure java.util.logging.

Configuring Logging
To generate a useful and detailed log, you must configure java.util.logging. This can be done either through a configuration file or programmatically. A sample configuration file, OracleLog.properties, is provided as part of the JDBC installation in the demo directory. It contains basic information about how to configure java.util.logging and provides some initial settings that you can start with. You may use this sample file as is, edit the file and use it, rename the file and edit it, or create an entirely new file of any name. To use a configuration file, you must identify it to the Java run-time. This can be done by setting a system property. For example:
java -Djava.util.logging.config.file=/jdbc/demo/OracleLog.properties.

It is read by the java.util.logging system. This file can reside anywhere. You can use both java.util.logging.config.file and oracle.jdbc.Trace at the same time.
java -Djava.util.logging.config.file=/jdbc/demo/OracleLog.properties -Doracle.jdbc.Trace=true

Diagnosability in JDBC

30-3

Logging

You can use the default OracleLog.properties file. It may or may not get you the desired output. You can also create and use your own configuration file by following these steps:
1. 2.

Create a file named myConfig.properties. You can use any name that you wish to. Insert the following lines of text in the file:
.level=SEVERE oracle.jdbc.level=INFO oracle.jdbc.handlers=java.util.logging.ConsoleHandler java.util.logging.ConsoleHandler.level=INFO java.util.logging.ConsoleHandler.formatter=java.util.logging.SimpleFormatter

3. 4.

Save the file. Set the system property to use this configuration file.
java -Djava.util.logging.config.file=<filepath>/myConfig.properties ...

where filepath is the path of the folder where you have saved the myConfig.properties file. You can also configure java.util.logging to dump the log output into a file. To do so, modify the configuration file as follows:
.level=SEVERE oracle.jdbc.level=INFO oracle.jdbc.handlers=java.util.logging.FileHandler java.util.logging.FileHandler.level=INFO java.util.logging.FileHandler.pattern=jdbc.log java.util.logging.FileHandler.count=1 java.util.logging.FileHandler.formatter=java.util.logging.SimpleFormatter

This will generate exactly the same log output and save it in a file named jdbc.log in the current directory. You can control the amount of detail by changing the level settings. The defined levels from the least detailed to the most detailed are:

OFF Turns off logging.

SEVERE Logs SQLExceptions and internal errors.

WARNING Logs SQLWarnings and bad but not fatal internal conditions.

INFO Logs infrequent but significant events and errors. It produces a relatively low volume of log messages.

CONFIG Logs SQL strings that are executed.

FINE Logs the entry and exit to every public method providing a detailed trace of JDBC operations. It produces a fairly high volume of log messages.

30-4 Oracle Database JDBC Developer's Guide and Reference

Logging

FINER Logs calls to internal methods.

FINEST Logs calls to high volume internal methods.

ALL Logs all the details. This is the most detailed level of logging.
Note:

Levels more detailed than FINE generate huge log volumes.

In the example provided earlier, to reduce the amount of detail, change the java.util.logging.FileHandler.level setting from ALL to INFO:
java.util.logging.FileHandler.level=INFO

Although you can, it is not necessary to change the level of the oracle.jdbc logger. Setting the FileHandler level will control what log messages are dumped into the log file.

Using Loggers
Setting the level reduces all the logging output from JDBC. However, sometimes you need a lot of output from one part of the code and very little from other parts. To do that you must understand more about loggers. Loggers exist in a tree structure defined by their names. The root logger is named "", the empty string. If you look at the first line of the configuration file you see .level=SEVERE. This is setting the level of the root logger. The next line is oracle.jdbc.level=INFO. This sets the level of the logger named oracle.jdbc. The oracle.jdbc logger is a member of the logger tree. Its parent is named oracle. The parent of the oracle logger is the root logger (the empty string). Logging messages are sent to a particular logger, for example, oracle.jdbc. If the message passes the level check at that level, then the message is passed to the handler at that level, if any, and to the parent logger. So a log message sent to oracle.log is compared against that logger's level, INFO if you are following along. If the level is the same or less (less detailed) then it is sent to the FileHandler and to the parent logger, 'oracle'. Again it is checked against the level. If as in this case, the level is not set then it uses the parent level, SEVERE. If the message level is the same or less it is passed to the handler, which there isn't one, and sent to the parent. In this case the parent in the root logger. All this tree stuff didn't help you reduce the amount of output. What will help is that the JDBC drivers use several subloggers. If you restrict the log messages to one of the subloggers you will get substantially less output. The loggers used by Oracle JDBC drivers include:

oracle.jdbc oracle.jdbc.driver oracle.jdbc.pool oracle.jdbc.rowset oracle.jdbc.xa oracle.sql

Diagnosability in JDBC

30-5

Logging

Note:

The loggers used by the drivers may vary from release to

release.

An Example
Suppose you want to trace what is happening in the oracle.sql component and also want to capture some basic information about the rest of the driver. This is a more complex use of logging. The following are the entries in the config file:
# # set levels # .level=SEVERE oracle.level=INFO oracle.jdbc.driver.level=INFO oracle.jdbc.pool.level=OFF oracle.jdbc.util.level=OFF oracle.sql.level=INFO # # configure handlers # oracle.handlers=java.util.logging.ConsoleHandler java.util.logging.ConsoleHandler.level=INFO java.util.logging.ConsoleHandler.formatter=java.util.logging.SimpleFormatter

Let us consider what each line in the configuration file is doing.

.level=SEVERE

Sets the logging level of the root logger to SEVERE. We do not want to see any logging from other, non-Oracle components unless something fails badly. Therefore, we set the default level for all loggers to SEVERE. Each logger inherits its level from its parent unless set explicitly. By setting the level of the root logger to SEVERE we ensure that all other loggers inherit that level except for the ones we set otherwise.
oracle.level=INFO

We want log output from both the oracle.sql and oracle.jdbc.driver loggers. Their common ancestor is oracle. Therefore, we set the level of the oracle logger to INFO. We will control the detail more explicitly at lower levels.
oracle.jdbc.driver.level=INFO

We only want to see the SQL execution from oracle.jdbc.driver. Therefore, we set the level to INFO. This is a fairly low volume level, but will help us to keep track of what our test is doing.
oracle.jdbc.pool.level=OFF

We are using a DataSource in our test and do not want to see all of that logging. Therefore, we turn it OFF.
oracle.jdbc.util.level=OFF

We do not want to see the logging from the oracle.jdbc.util package. If we were using XA or rowsets we would turn them off as well.
oracle.sql.level=INFO

30-6 Oracle Database JDBC Developer's Guide and Reference

Logging

We want to see what is happening in oracle.sql. Therefore, we set the level to INFO. This provides a lot of information about the public method calls without overwhelming detail.
oracle.handlers=java.util.logging.ConsoleHandler

We are going to dump everything to stderr. When we run the test we will redirect stderr to a file.
java.util.logging.ConsoleHandler.level=INFO

We want to dump everything to the console which is System.err. In this case, we are doing the filtering with the loggers rather than the handler.
java.util.logging.ConsoleHandler.formatter=java.util.logging.SimpleFormatter

We will use a simple, more or less human readable format. When you run your test with this configuration file, you will get moderately detailed information from the oracle.sql package, a little bit of information from the core driver code, and nothing from any other code. You can also use XMLFormatter. XMLFormatter is the best choice for logs that you send to Oracle Support, because it enables us to more easily use automated processing of the log output.

Performance, Scalability, and Security Issues

Although the logging feature enables you to trace or debug your application and generate detail log output, it has certain performance, scalability, and security issues. Performance and Scalability Issues Logging has substantial impact on performance. However, JDBC logging is generally not enabled in production systems. When logging is disabled, it will have no impact on performance. It also has a negative impact on scalability. Logging involves protected access to a number of shared resources resulting in severely reduced scalability. This is intrinsic to the java.util.logging framework. However, in a typical production system, JDBC logging is not enabled and, therefore, will not have an impact on scalability. Security Concerns When full logging is enabled, it is almost guaranteed that all sensitive information will be exposed in the log files. This is intrinsic to the logging feature. However, only certain JDBC JAR files include the JDBC logging feature. The following JAR files include full logging and should not be used in a sensitive environment:

ojdbc5_g.jar ojdbc5dms_g.jar ojdbc6_g.jar ojdbc6dms_g.jar

The following JAR files include a limited logging capability:

ojdbc5dms.jar ojdbc6dms.jar

Diagnosability in JDBC

30-7

Diagnosability Management

Note:

Database user names and passwords do not appear in log files created by these JAR files. However, sensitive user data that is part of a SQL statement, a defined value, or a bind value can appear in a log created using one of these JAR files.

Diagnosability Management
The JDBC diagnosability management feature introduces an MBean, oracle.jdbc.driver.OracleDiagnosabilityMBean. This MBean provides means to enable and disable JDBC logging.
See Also: Refer to the JDBC JavaDoc for information about the OracleDiagnosabilityMBean API.

In future releases, the MBean will be enhanced to provide additional statistics about JDBC internals. Security Concerns This feature can enable JDBC logging. Enabling JDBC logging does not require any special permission. However, once logging is enabled, generating any log output requires the standard Java permission LoggingPermission. Without this permission, any JDBC operation that would generate log output will throw a security exception. This is a standard Java mechanism.

30-8 Oracle Database JDBC Developer's Guide and Reference

31
JDBC DMS Metrics
There are two kinds of metrics, end-to-end metrics and Dynamic Monitoring Service (DMS) metrics. End-to-end metrics are used for tagging application activity from the entry into the application code through JDBC to the database and back. DMS metrics are used to measure the performance of application components. Customer use of end-to-end metrics in JDBC is generally discouraged. This chapter discusses the DMS metrics generated by Oracle JDBC 11.1 in following sections:

Overview Determining the Type of Metric to be Generated Generating the SQLText Metric Accessing DMS Metrics Using JMX

Overview
Starting with Oracle Database 10g, Oracle Java Database Connectivity (JDBC) supports end-to-end metrics. In Oracle Database 11g Release 1 (11.1), the end-to-end metrics can be set directly by an application only when it is not using a DMS-enabled JAR. But, if your application is using a DMS-enabled JAR, the end-to-end metrics can be set only through DMS. All DMS metrics are available in the following DMS-enabled JARs:

ojdbc15dms.jar ojdbc15dms_g.jar ojdbc6dms.jar ojdbc6dms_g.jar

Any other JDBC JAR files do not generate any DMS metrics. The metrics generated in Oracle JDBC 11.1 release are different from 10.2, 10.1, 9.2, and earlier versions of Oracle JDBC as it makes no attempt to retain compatibility with earlier versions. There are also no compatibility modes. A system that is dependent on the exact details of the DMS metrics generated by earlier versions of JDBC may have unexpected behavior when processing the metrics generated by Oracle JDBC 11.1. This is by design and cannot be changed. Statement metrics can be reported consolidated for all statements in a connection or individually for each statement. All DMS metrics, except those related to individual statements, are enabled at all times.

JDBC DMS Metrics 31-1

Determining the Type of Metric to be Generated

Note:

You can enable or disable the SQLText statement metric. It is disabled by default. If enabled, it is enabled for all statements.

Determining the Type of Metric to be Generated

To determine whether to use consolidated or individual metrics, JDBC checks the DMSConsole sensor weight. If the sensor weight is less than or equal to DMSConsole.NORMAL, then JDBC generates consolidated statement metrics. If the sensor weight is greater than DMSConsole.NORMAL, then JDBC generates individual statement metrics. JDBC checks the DMSConsole sensor weight when creating a Prepared or Callable statement and depending on the sensor weight at the time the statement is created, the metrics are generated. Changing the value of the sensor weight, after the statement has been created, does not cause a statement to switch between consolidated and individual metrics.
Note:

In the presence of statement caching, it may appear that changing sensor weight has no impact as statements are retrieved from the cache rather than created anew.

There is only one list of statement metrics that is generated for both consolidated and individual statement metrics. The only difference between these two lists is the aggregation of the statements. When individual statement metrics are generated, one set of metrics is generated for each distinct statement object created by the JDBC driver. On the other hand, when consolidated statement metrics are generated, all statements created by a given connection use the same set of statement metrics. For example, consider an 'execute' phase event. If individual statement metrics are used, then each statement created will have a distinct 'execute' phase event. So, from two such statements, it will be possible to distinguish the execution statistics for the two separate statements. If one has an execution time of 1 second and the other an execution time of 3 seconds, then there will be two distinct 'execute' phase events, one with a total time and average of 1 second and the other with a total time and average of 3 seconds. But, if consolidated statement metrics are used, all statements will use the single 'execute' phase event common to the connection. So, from two such statements created by the same connection, it will not be possible to distinguish the execution statistics for the two statements. If one has an execution time of 1 second and the other an execution time of 3 seconds, then the common 'execute' phase event will report a total execution time of 4 seconds and an average of 2 seconds.

Generating the SQLText Metric

Depending on the version of DMS, there are two mechanisms for determining the generating of the SQLText statement metrics:

If the 11.1 version of the DMS JAR file is present in the classpath environment variable, then JDBC checks the DMS update SQL text flag. If this flag is true, then the SQLText metric is updated. If the 11.1 version of the DMS JAR file is not present in the classpath environment variable, then JDBC uses the value of the DMSStatementMetrics connection property. If this statement property is true, then SQLText metric is updated. The default value of this connection property is false.

31-2 Oracle Database JDBC Developer's Guide and Reference

Accessing DMS Metrics Using JMX

Whether SQLText metric will be generated or not, is independent on the use of the type of statement metrics used, that is, individual statement metrics or consolidated statement metrics

Accessing DMS Metrics Using JMX

JMX (Java Management Extensions) is a Java technology that supplies tools for managing and monitoring applications, system objects, devices, service-oriented networks, and the JVM (Java Virtual Machine). You can easily access DMS metrics at run time using a management application that supports JMX. For more information about using JMX to access DMS data, visit the following link: http://www.oracle.com/technology/products/ias/toplink/doc/1013/m ain/_html/optimiz004.htm#BEEFGGBE

JDBC DMS Metrics 31-3

Accessing DMS Metrics Using JMX

31-4 Oracle Database JDBC Developer's Guide and Reference

Part IX
Appendixes
This part consists of appendixes that discuss Java Database Connectivity (JDBC) reference information, tips for coding JDBC applications, JDBC error messages, and troubleshooting JDBC applications. Part IX contains the following appendixes:

Appendix A, "Reference Information" Appendix B, "Coding Tips" Appendix C, "JDBC Error Messages" Appendix D, "Troubleshooting"

A
Reference Information
This appendix contains detailed Java Database Connectivity (JDBC) reference information, including the following topics:

Valid SQL-JDBC Data Type Mappings Supported SQL and PL/SQL Data Types Embedded SQL92 Syntax Oracle JDBC Notes and Limitations

Valid SQL-JDBC Data Type Mappings

Table 111 describes the default mappings between Java classes and SQL data types supported by Oracle JDBC drivers. Compare the contents of the JDBC Type Codes, Standard Java Types, and SQL Data Types columns in Table 111 with the contents of Table A1. Table A1 lists all the possible Java types to which a given SQL data type can be validly mapped. Oracle JDBC drivers will support these nondefault mappings. For example, to materialize SQL CHAR data in an oracle.sql.CHAR object, use the getCHAR method. To materialize it as a java.math.BigDecimal object, use the getBigDecimal method.
Note:

For classes where oracle.sql.ORAData appears in italic, these can be generated by JPublisher.

Table A1

Valid SQL Data Type-Java Class Mappings Can be materialized as these Java types: oracle.sql.CHAR java.lang.String java.sql.Date java.sql.Time java.sql.Timestamp java.lang.Byte java.lang.Short java.lang.Integer java.lang.Long

These SQL data types: CHAR, VARCHAR2, LONG

Reference Information A-1

Valid SQL-JDBC Data Type Mappings

Table A1 (Cont.) Valid SQL Data Type-Java Class Mappings These SQL data types: Can be materialized as these Java types: java.lang.Float java.lang.Double java.math.BigDecimal byte, short, int, long, float, double DATE oracle.sql.DATE java.sql.Date java.sql.Time java.sql.Timestamp java.lang.String NUMBER oracle.sql.NUMBER java.lang.Byte java.lang.Short java.lang.Integer java.lang.Long java.lang.Float java.lang.Double java.math.BigDecimal byte, short, int, long, float, double OPAQUE RAW, LONG RAW oracle.sql.OPAQUE oracle.sql.RAW byte[] ROWID oracle.sql.CHAR oracle.sql.ROWID java.lang.String BFILE BLOB oracle.sql.BFILE oracle.sql.BLOB java.sql.Blob CLOB oracle.sql.CLOB java.sql.Clob TIMESTAMP java.sql.Date, oracle.sql.DATE, java.sql.Time, java.sql.Timestamp, oracle.sql.TIMESTAMP, java.lang.String, byte[] java.sql.Date, oracle.sql.DATE, java.sql.Time, java.sql.Timestamp, oracle.sql.TIMESTAMPTZ, java.lang.String, byte[] java.sql.Date, oracle.sql.DATE, java.sql.Time, java.sql.Timestamp, oracle.sql.TIMESTAMPLTZ, java.lang.String, byte[]

TIMESTAMP WITH TIME ZONE

TIMESTAMP WITH LOCAL TIME ZONE

A-2 Oracle Database JDBC Developer's Guide and Reference

Supported SQL and PL/SQL Data Types

Table A1 (Cont.) Valid SQL Data Type-Java Class Mappings These SQL data types: Object types Can be materialized as these Java types: oracle.sql.STRUCT java.sql.Struct java.sql.SqlData oracle.sql.ORAData Reference types oracle.sql.REF java.sql.Ref oracle.sql.ORAData Nested table types and VARRAY types oracle.sql.ARRAY java.sql.Array oracle.sql.ORAData

Note:

The type UROWID is not supported. The oracle.sql.Datum class is abstract. The value passed to a parameter of type oracle.sql.Datum must be of the Java type corresponding to the underlying SQL type. Likewise, the value returned by a method with return type oracle.sql.Datum must be of the Java type corresponding to the underlying SQL type.

Supported SQL and PL/SQL Data Types

The tables in this section list SQL and PL/SQL data types, and whether Oracle JDBC drivers support them. Table A2 describes Oracle JDBC driver support for SQL data types.
Table A2 Support for SQL Data Types Supported by JDBC Drivers? yes yes yes yes yes no (see Note) no yes no (see Note) yes yes

SQL Data Type BFILE BLOB CHAR CLOB DATE NCHAR NCHAR VARYING NUMBER NVARCHAR2 RAW REF

Reference Information A-3

Supported SQL and PL/SQL Data Types

Table A2 (Cont.) Support for SQL Data Types SQL Data Type ROWID UROWID VARCHAR2 Supported by JDBC Drivers? yes no yes

Note: The types NCHAR and NVARCHAR2 are supported indirectly. There is no corresponding java.sql.Types type, but if your application calls formOfUse(NCHAR), then these types can be accessed.

Table A3 describes Oracle JDBC support for the ANSI-supported SQL data types.
Table A3 Support for ANSI-92 SQL Data Types

ANSI-Supported SQL Data Type Supported by JDBC Drivers? CHARACTER DEC DECIMAL DOUBLE PRECISION FLOAT INT INTEGER NATIONAL CHARACTER NATIONAL CHARACTER VARYING NATIONAL CHAR NATIONAL CHAR VARYING NCHAR NCHAR VARYING NUMERIC REAL SMALLINT VARCHAR yes yes yes yes yes yes yes no no yes no yes no yes yes yes yes

Table A4 describes Oracle JDBC driver support for SQL User-Defined types.
Table A4 Support for SQL User-Defined Types Supported by JDBC Drivers? yes yes yes

SQL User-Defined type OPAQUE Reference types Object types (JAVA_OBJECT)

A-4 Oracle Database JDBC Developer's Guide and Reference

Supported SQL and PL/SQL Data Types

Table A4 (Cont.) Support for SQL User-Defined Types SQL User-Defined type Nested table types and VARRAY types Supported by JDBC Drivers? yes

Table A5 describes Oracle JDBC driver support for PL/SQL data types. Note that PL/SQL data types include these categories:

Scalar types Scalar character types, which includes BOOLEAN and DATE data types Composite types Reference types Large object (LOB) types
Support for PL/SQL Data Types Supported by JDBC Drivers?

Table A5

PL/SQL Data Type Scalar Types: BINARY INTEGER DEC DECIMAL DOUBLE PRECISION FLOAT INT INTEGER NATURAL NATURALn NUMBER NUMERIC PLS_INTEGER POSITIVE POSITIVEn REAL SIGNTYPE SMALLINT Scalar Character Types: CHAR CHARACTER LONG LONG RAW NCHAR NVARCHAR2

yes yes yes yes yes yes yes yes no yes yes yes yes no yes yes yes

yes yes yes yes no (see Note) no (see Note)

Reference Information A-5

Embedded SQL92 Syntax

Table A5 (Cont.) Support for PL/SQL Data Types PL/SQL Data Type RAW ROWID STRING UROWID VARCHAR VARCHAR2 BOOLEAN DATE Composite Types: RECORD TABLE VARRAY Reference Types: REF CURSOR types object reference types LOB Types: BFILE BLOB CLOB NCLOB yes yes yes yes yes yes no no yes Supported by JDBC Drivers? yes yes yes no yes yes yes yes

Note:

The types NATURAL, NATURALn, POSITIVE, POSITIVEn, and SIGNTYPE are subtypes of BINARY INTEGER. The types DEC, DECIMAL, DOUBLE PRECISION, FLOAT, INT, INTEGER, NUMERIC, REAL, and SMALLINT are subtypes of NUMBER. The types NCHAR and NVARCHAR2 are supported indirectly. There is no corresponding java.sql.Types type, but if your application calls formOfUse(NCHAR), then these types can be accessed. Refer to "NCHAR, NVARCHAR2, NCLOB and the defaultNChar Property in JDK 1.5" on page 19-3 for details.

Embedded SQL92 Syntax

Oracle JDBC drivers support some embedded SQL92 syntax, which is the syntax that you specify between curly braces. The current support is basic. This section describes the support offered by the drivers for the following SQL92 constructs:

Time and Date Literals Scalar Functions

A-6 Oracle Database JDBC Developer's Guide and Reference

Embedded SQL92 Syntax

LIKE Escape Characters Outer Joins Function Call Syntax

Where driver support is limited, these sections also describe possible workarounds. Disabling Escape Processing Escape processing for SQL92 syntax is enabled by default, which results in the JDBC driver performing escape substitution before sending the SQL code to the database. If you want the driver to use regular Oracle SQL syntax, which is more efficient than SQL92 syntax and escape processing, then use this statement:
stmt.setEscapeProcessing(false);

Time and Date Literals

Databases differ in the syntax they use for date, time, and timestamp literals. JDBC supports dates and times written only in a specific format. This section describes the formats you must use for date, time, and timestamp literals in SQL statements.

Date Literals
The JDBC drivers support date literals in SQL statements written in the format:
{d 'yyyy-mm-dd'}

Where yyyy-mm-dd represents the year, month, and day. For example:
{d '1995-10-22'}

The JDBC drivers will replace this escape clause with the equivalent Oracle representation: "22 OCT 1995". The following code snippet contains an example of using a date literal in a SQL statement.
// Connect to the database // You can put a database name after the @ sign in the connection URL. OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:oracle:oci:@"); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection(); // Create a Statement Statement stmt = conn.createStatement (); // Select the ename column from the emp table where the hiredate is Jan-23-1982 ResultSet rset = stmt.executeQuery ("SELECT ename FROM emp WHERE hiredate = {d '1982-01-23'}"); // Iterate through the result and print the employee names while (rset.next ()) System.out.println (rset.getString (1));

Time Literals
The JDBC drivers support time literals in SQL statements written in the format:
Reference Information A-7

Embedded SQL92 Syntax

{t 'hh:mm:ss'}

where, hh:mm:ss represents the hours, minutes, and seconds. For example:
{t '05:10:45'}

The JDBC drivers will replace this escape clause with the equivalent Oracle representation: "05:10:45". If the time is specified as:
{t '14:20:50'}

Then the equivalent Oracle representation would be "14:20:50", assuming the server is using a 24-hour clock. This code snippet contains an example of using a time literal in a SQL statement.
ResultSet rset = stmt.executeQuery ("SELECT ename FROM emp WHERE hiredate = {t '12:00:00'}");

Timestamp Literals
The JDBC drivers support timestamp literals in SQL statements written in the format:
{ts 'yyyy-mm-dd hh:mm:ss.f...'}

where yyyy-mm-dd hh:mm:ss.f... represents the year, month, day, hours, minutes, and seconds. The fractional seconds portion (.f...) is optional and can be omitted. For example: {ts '1997-11-01 13:22:45'} represents, in Oracle format, NOV 01 1997 13:22:45. This code snippet contains an example of using a timestamp literal in a SQL statement.
ResultSet rset = stmt.executeQuery ("SELECT ename FROM emp WHERE hiredate = {ts '1982-01-23 12:00:00'}");

Mapping SQL DATE Data type to Java Oracle Database 8i and earlier versions did not support TIMESTAMP data, but Oracle DATE data used to have a time component as an extension to the SQL standard. So, Oracle Database 8i and earlier versions of JDBC drivers mapped oracle.sql.DATE to java.sql.Timestamp to preserve the time component. Starting with Oracle Database 9.0.1, TIMESTAMP support was included and 9i JDBC drivers started mapping oracle.sql.DATE to java.sql.Date. This mapping was incorrect as it truncated the time component of Oracle DATE data. To overcome this problem, Oracle Database 11.1 introduces a new flag mapDateToTimestamp. The default value of this flag is true, which means that by default the drivers will correctly map oracle.sql.DATE to java.sql.Timestamp, retaining the time information. If you still want the incorrect but 10g compatible oracle.sql.DATE to java.sql.Date mapping, then you can get it by setting the value of mapDateToTimestamp flag to false.

A-8 Oracle Database JDBC Developer's Guide and Reference

Embedded SQL92 Syntax

Note: To overcome the problem of oracle.sql.DATE to java.sql.Date mapping, Oracle Database 9.2 had introduced a flag, V8Compatible. The default value of this flag was false, which allowed the mapping of Oracle DATE data to java.sql.Date data. But, users could retain the time component of the Oracle DATE data by setting the value of this flag to true. This flag is deprecated in 11g because it controlled Oracle Database 8i compatibility, which is no longer supported.

Scalar Functions
Oracle JDBC drivers do not support all scalar functions. To find out which functions the drivers support, use the following methods supported by the Oracle-specific oracle.jdbc.OracleDatabaseMetaData class and the standard Java java.sql.DatabaseMetadata interface:

getNumericFunctions() Returns a comma-delimited list of math functions supported by the driver. For example, ABS, COS, SQRT.

getStringFunctions() Returns a comma-delimited list of string functions supported by the driver. For example, ASCII, LOCATE.

getSystemFunctions() Returns a comma-delimited list of system functions supported by the driver. For example, DATABASE, USER.

getTimeDateFunctions() Returns a comma-delimited list of time and date functions supported by the driver. For example, CURDATE, DAYOFYEAR, HOUR.
Note:

Oracle JDBC drivers support fn, the function keyword.

LIKE Escape Characters

The characters % and _ have special meaning in SQL LIKE clauses. You use % to match zero or more characters and _ to match exactly one character. If you want to interpret these characters literally in strings, then you precede them with a special escape character. For example, if you want to use ampersand (&) as the escape character, then you identify it in the SQL statement as:
Statement stmt = conn.createStatement (); // Select the empno column from the emp table where the ename starts with '_' ResultSet rset = stmt.executeQuery ("SELECT empno FROM emp WHERE ename LIKE '&_%' {ESCAPE '&'}"); // Iterate through the result and print the employee numbers while (rset.next ()) System.out.println (rset.getString (1));

Reference Information A-9

Embedded SQL92 Syntax

Note:

If you want to use the backslash character (\) as an escape character, then you must enter it twice, that is, \\. For example:

ResultSet rset = stmt.executeQuery("SELECT empno FROM emp WHERE ename LIKE '\\_%' {escape '\\'}");

Outer Joins
Oracle JDBC drivers do not support the outer join syntax. The workaround is to use Oracle outer join syntax: Instead of:
Statement stmt = conn.createStatement (); ResultSet rset = stmt.executeQuery ("SELECT ename, dname FROM {OJ dept LEFT OUTER JOIN emp ON dept.deptno = emp.deptno} ORDER BY ename");

Use Oracle SQL syntax:

Statement stmt = conn.createStatement (); ResultSet rset = stmt.executeQuery ("SELECT ename, dname FROM emp a, dept b WHERE a.deptno = b.deptno(+) ORDER BY ename");

Function Call Syntax

Oracle JDBC drivers support the following procedure and function call syntax: Procedure calls:
{ call procedure_name (argument1, argument2,...) }

Function calls:
{ ? = call procedure_name (argument1, argument2,...) }

SQL92 to SQL Syntax Example

You can write a simple program to translate SQL92 syntax to standard SQL syntax. The following program prints the comparable SQL syntax for SQL92 statements for function calls, date literals, time literals, and timestamp literals. In the program, the oracle.jdbc.OracleSql class parse() method performs the conversions.
public class Foo { static oracle.jdbc.OracleDriver driver = new oracle.jdbc.OracleDriver(); public static void main (String args[]) throws Exception { show ("{call foo(?, ?)}"); show ("{? = call bar (?, ?)}"); show ("{d '1998-10-22'}"); show ("{t '16:22:34'}"); show ("{ts '1998-10-22 16:22:34'}"); }

A-10 Oracle Database JDBC Developer's Guide and Reference

Oracle JDBC Notes and Limitations

public static void show (String s) throws Exception { System.out.println (s + " => " + driver.processSqlEscapes(s)); } }

The following code is the output that prints the comparable SQL syntax.
{call foo(?, ?)} => BEGIN foo(:1, :2); END; {? = call bar (?, ?)} => BEGIN :1 := bar (:2, :3); END; {d '1998-10-22'} => TO_DATE ('1998-10-22', 'YYYY-MM-DD') {t '16:22:34'} => TO_DATE ('16:22:34', 'HH24:MI:SS') {ts '1998-10-22 16:22:34'} => TO_TIMESTAMP ('1998-10-22 16:22:34', 'YYYY-MM-DD HH24:MI:SS.FF')

Oracle JDBC Notes and Limitations

The following limitations exist in the Oracle JDBC implementation, but all of them are either insignificant or have easy workarounds. This section covers the following topics:

CursorName SQL92 Outer Join Escapes PL/SQL TABLE, BOOLEAN, and RECORD Types IEEE 754 Floating Point Compliance Catalog Arguments to DatabaseMetaData Calls SQLWarning Class Binding Named Parameters

CursorName
Oracle JDBC drivers do not support the get getCursorName and setCursorName methods, because there is no convenient way to map them to Oracle constructs. Oracle recommends using ROWID instead.
See Also: "Oracle ROWID Type" on page 4-13 for more information about how to use and manipulate ROWIDs.

SQL92 Outer Join Escapes

Oracle JDBC drivers do not support SQL92 outer join escapes. Use Oracle SQL syntax with + instead.
See Also:

"Embedded SQL92 Syntax" on page A-6

PL/SQL TABLE, BOOLEAN, and RECORD Types

It is not feasible for Oracle JDBC drivers to support calling arguments or return values of the PL/SQL RECORD, BOOLEAN, or table with non-scalar element types. However, Oracle JDBC drivers support PL/SQL index-by table of scalar element types.
See Also:

"Accessing PL/SQL Index-by Tables" on page 4-4

Reference Information

A-11

Oracle JDBC Notes and Limitations

As a workaround to PL/SQL RECORD, BOOLEAN, or non-scalar table types, create container procedures that handle the data as types supported by JDBC. For example, to wrap a stored procedure that uses PL/SQL boolean, create a stored procedure that takes a character or number from JDBC and passes it to the original procedure as BOOLEAN or, for an output parameter, accepts a BOOLEAN argument from the original procedure and passes it as a CHAR or NUMBER to JDBC. Similarly, to wrap a stored procedure that uses PL/SQL records, create a stored procedure that handles a record in its individual components, such as CHAR and NUMBER, or in a structured object type. To wrap a stored procedure that uses PL/SQL tables, break the data into components or perhaps use Oracle collection types.
See Also: "Boolean Parameters in PL/SQL Stored Procedures" on page D-2 for an example of a workaround for BOOLEAN.

IEEE 754 Floating Point Compliance

The arithmetic for the Oracle NUMBER type does not comply with the IEEE 754 standard for floating-point arithmetic. Therefore, there can be small disagreements between the results of computations performed by Oracle and the same computations performed by Java. Oracle stores numbers in a format compatible with decimal arithmetic and guarantees 38 decimal digits of precision. It represents zero, minus infinity, and plus infinity exactly. For each positive number it represents, it represents a negative number of the same absolute value. It represents every positive number between 10-30 and (1 10-38) * 10126 to full 38-digit precision.

Catalog Arguments to DatabaseMetaData Calls

Certain DatabaseMetaData methods define a catalog parameter. This parameter is one of the selection criteria for the method. Oracle does not have multiple catalogs, but it does have packages.
See Also: "DatabaseMetaData TABLE_REMARKS Reporting" on page 23-19 for information about how Oracle JDBC drivers treat the catalog argument.

SQLWarning Class
The java.sql.SQLWarning class provides information about a database access warning. Warnings typically contain a description of the warning and a code that identifies the warning. Warnings are silently chained to the object whose method caused it to be reported. Oracle JDBC drivers generally do not support SQLWarning. As an exception to this, scrollable result set operations do generate SQL warnings, but the SQLWarning instance is created on the client, not in the database.
See Also:

"Processing SQL Exceptions" on page 2-14

Binding Named Parameters

Binding by name is not supported when using the setXXX methods. Under certain circumstances, previous versions of Oracle JDBC drivers have allowed binding statement variables by name when using the setXXX methods. In the following statement, the named variable EmpId would be bound to the integer 314159.
PreparedStatement p = conn.prepareStatement

A-12 Oracle Database JDBC Developer's Guide and Reference

Oracle JDBC Notes and Limitations

("SELECT name FROM emp WHERE id = :EmpId"); p.setInt(1, 314159);

This capability to bind by name using the setXXX methods is not part of the JDBC specification, and Oracle does not support it. The JDBC drivers can throw a SQLException or produce unexpected results. In Oracle Database 10g JDBC drivers, bind by name is supported using the setXXXAtName methods.
See Also: "Interface oracle.jdbc.OracleCallableStatement" on page 4-22 and "Interface oracle.jdbc.OraclePreparedStatement" on page 4-22

The bound values are not copied by the drivers until you call the execute method. So, changing the bound value before calling the execute method could change the bound value. For example, consider the folllowing code snippet:
PreparedStatement p; ....... Date d = new Date(1181676033917L); p.setDate(1, d); d.setTime(0); p.executeUpdate();

This code snippet inserts Date(0) in the database instead of Date(1181676033917L) because the bound values are not copied by JDBC driver implementation for performance reasons. Retaining Bound Values Before Oracle9i Database, Oracle JDBC drivers did not retain bound values from one call of execute to the next as specified in JDBC 1.0. All releases after Oracle9i Database have retained bound values. For example:
PreparedStatement p = conn.prepareStatement ("SELECT name FROM emp WHERE id = ? AND dept = ?"); p.setInt(1, 314159); p.setString(2, "SALES"); ResultSet r1 = p.execute(); p.setInt(1, 425260); ResultSet r2 = p.execute();

Previously, a SQLException would be thrown by the second execute call because no value was bound to the second argument. In Oracle Database 10g, the second execute will return the correct value, retaining the binding of the second argument to the string SALES. If the retained bound value is a stream, then Oracle JDBC drivers will not reset the stream. Unless the application code resets, repositions, or otherwise modifies the stream, the subsequent calls to execute will send NULL as the value of the argument.

Reference Information

A-13

Oracle JDBC Notes and Limitations

A-14 Oracle Database JDBC Developer's Guide and Reference

B
Coding Tips
This appendix describes methods to optimize a Java Database Connectivity (JDBC) application or applet. It includes the following topics:

JDBC and Multithreading Performance Optimization Transaction Isolation Levels and Access Modes

JDBC and Multithreading

Oracle JDBC drivers provide full support for, and are highly optimized for, applications that use Java multithreading. Controlled serial access to a connection, such as that provided by connection caching, is both necessary and encouraged. However, Oracle strongly discourages sharing a database connection among multiple threads. Avoid allowing multiple threads to access a connection simultaneously. If multiple threads must share a connection, use a disciplined begin-using/end-using protocol.

Performance Optimization
You can significantly enhance the performance of your JDBC programs by using any of these features:

Disabling Auto-Commit Mode Standard Fetch Size and Oracle Row Prefetching Standard and Oracle Update Batching Statement Caching Mapping Between Built-in SQL and Java Types

Disabling Auto-Commit Mode

Auto-commit mode indicates to the database whether to issue an automatic COMMIT operation after every SQL operation. Being in auto-commit mode can be expensive in terms of time and processing effort if, for example, you are repeating the same statement with different bind variables. By default, new connection objects are in auto-commit mode. However, you can disable auto-commit mode with the setAutoCommit method of the connection object, either java.sql.Conection or oracle.jdbc.OracleConnection.

Coding Tips

B-1

Performance Optimization

In auto-commit mode, the COMMIT operation occurs either when the statement completes or the next execute occurs, whichever comes first. In the case of statements returning a ResultSet, the statement completes when the last row of the ResultSet has been retrieved or when the ResultSet has been closed. In more complex cases, a single statement can return multiple results as well as output parameter values. Here, the COMMIT occurs when all results and output parameter values have been retrieved. If you disable auto-commit mode with a setAutoCommit(false) call, then you must manually commit or roll back groups of operations using the commit or rollback method of the connection object. Example The following example illustrates loading the driver and connecting to the database. Because new connections are in auto-commit mode by default, this example shows how to disable auto-commit. In the example, conn represents the Connection object, and stmt represents the Statement object.
// Connect to the database // You can put a database host name after the @ sign in the connection URL. OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:oracle:oci:@"); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection(); // It's faster when auto commit is off conn.setAutoCommit (false); // Create a Statement Statement stmt = conn.createStatement (); ...

Standard Fetch Size and Oracle Row Prefetching

Oracle JDBC connection and statement objects allow you to specify the number of rows to prefetch into the client with each trip to the database while a result set is being populated during a query. You can set a value in a connection object that affects each statement produced through that connection, and you can override that value in any particular statement object. The default value in a connection object is 10. Prefetching data into the client reduces the number of round-trips to the server. Similarly, and with more flexibility, JDBC 2.0 enables you to specify the number of rows to fetch with each trip, both for statement objects (affecting subsequent queries) and for result set objects (affecting row refetches). By default, a result set uses the value for the statement object that produced it. If you do not set the JDBC 2.0 fetch size, then the Oracle connection row-prefetch value is used by default.
See Also:

"Fetch Size" on page 17-15

Standard and Oracle Update Batching

Oracle JDBC drivers allow you to accumulate INSERT, DELETE, and UPDATE operations of prepared statements at the client and send them to the server in batches. This feature reduces round-trips to the server. You can either use Oracle update batching, which typically processes a batch implicitly once a pre-set batch value is reached, or standard update batching, where the batch is processed explicitly.

B-2 Oracle Database JDBC Developer's Guide and Reference

Performance Optimization

See Also:

"Update Batching" on page 23-1

Statement Caching
Statement caching improves performance by caching executable statements that are used repeatedly, such as in a loop or in a method that is called repeatedly. Applications use the statement cache to cache statements associated with a particular physical connection. When you enable statement caching, a statement object is cached when you call the close method. Because each physical connection has its own cache, multiple caches can exist if you enable statement caching for multiple physical connections. When you enable statement caching on a connection cache, the logical connections benefit from the statement caching that is enabled on the underlying physical connection. If you try to enable statement caching on a logical connection held by a connection cache, then this will throw an exception.
See Also:

Chapter 20, "Statement and Resultset Caching"

Mapping Between Built-in SQL and Java Types

The SQL built-in types are those types with system-defined names, such as NUMBER, and CHAR, as opposed to the Oracle objects, varray, and nested table types, which have user-defined names. In JDBC programs that access data of built-in SQL types, all type conversions are unambiguous, because the program context determines the Java type to which a SQL datum will be converted. Table B1 is a subset of the information presented in Table 111. The table lists the one-to-one type-mapping of the SQL database type to its Java oracle.sql.* representation.
Table B1 Mapping of SQL Data Types to Java Classes that Represent SQL Data Types ORACLE Mapping - Java Classes Representing SQL Data Types oracle.sql.CHAR oracle.sql.CHAR oracle.sql.DATE oracle.sql.NUMBER oracle.sql.NUMBER oracle.sql.NUMBER oracle.sql.NUMBER oracle.sql.NUMBER oracle.sql.RAW oracle.sql.RAW java.sql.ResultSet oracle.sql.CLOB oracle.sql.BLOB oracle.sql.BFILE oracle.sql.ARRAY oracle.sql.ARRAY

SQL Data Type CHAR VARCHAR2 DATE DECIMAL DOUBLE PRECISION FLOAT INTEGER REAL RAW LONG RAW REF CURSOR CLOB LOCATOR BLOB LOCATOR BFILE nested table varray

Coding Tips

B-3

Transaction Isolation Levels and Access Modes

Table B1 (Cont.) Mapping of SQL Data Types to Java Classes that Represent SQL Data SQL Data Type SQL object value ORACLE Mapping - Java Classes Representing SQL Data Types If there is no entry for the object value in the type map:

oracle.sql.STRUCT customized Java class

If there is an entry for the object value in the type map:

REF to SQL object type

class that implements oracle.sql.SQLRef, typically by extending oracle.sql.REF

This mapping provides the most efficient conversion between SQL and Java data representations. It stores the usual representations of SQL data as byte arrays. It avoids re-formatting the data. It is information-preserving. This Oracle mapping is the most efficient type-mapping for applications that "shovel" data from SQL to Java, or vice versa. The most efficient way to access numeric data is as primitive Java types like int, float, long, and double. However, the range of values of these types do not exactly match the range of values of the SQL NUMBER data type. As a result, there may be some loss of information. All character data is converted to the UCS2 character set of Java. The most efficient way to access character data is as java.lang.String. In worst case, this can cause a loss of information when two or more characters in the database character set map to a single UCS2 character. In Oracle Database 10g, all characters in the character set map to the characters in the UCS2 character set. However, some characters do map to surrogate pairs.

Transaction Isolation Levels and Access Modes

Read-only connections are supported by the Oracle server, but not by Oracle JDBC drivers. For transactions, the Oracle server supports only the TRANSACTION_READ_COMMITTED and TRANSACTION_SERIALIZABLE transaction isolation levels. The default is TRANSACTION_READ_COMMITTED. Use the following methods of the oracle.jdbc.OracleConnection interface to get and set the level:

getTransactionIsolation: Gets this connection's current transaction isolation level. setTransactionIsolation: Changes the transaction isolation level, using one of the TRANSACTION_* values.

B-4 Oracle Database JDBC Developer's Guide and Reference

C
JDBC Error Messages
This appendix briefly discusses the general structure of Java Database Connectivity (JDBC) error messages, then lists general JDBC error messages and TTC error messages that Oracle JDBC drivers can return. The appendix is organized as follows:

General Structure of JDBC Error Messages General JDBC Messages Native XA Messages TTC Messages

Each of the message lists is first sorted by ORA number, and then alphabetically.
See Also:

"Processing SQL Exceptions" on page 2-14

General Structure of JDBC Error Messages

The general JDBC error message structure allows run-time information to be appended to the end of a message, following a colon, as follows:
<error_message>:<extra_info>

For example, a "closed statement" error might be displayed as follows:

Closed Statement:next

This indicates that the exception was thrown during a call to the next method (of a result set object). In some cases, the user can find the same information in a stack trace.

General JDBC Messages

This section lists general JDBC error messages, first sorted by the ORA number, and then in alphabetic order in the following subsections:

JDBC Messages Sorted by ORA Number JDBC Messages Sorted in Alphabetic Order

JDBC Messages Sorted by ORA Number

The following table lists the JDBC error messages sorted by the ORA number:

JDBC Error Messages C-1

General JDBC Messages

ORA Number ORA-17001 ORA-17002 ORA-17003 ORA-17004 ORA-17005 ORA-17006 ORA-17007 ORA-17008 ORA-17009 ORA-17010 ORA-17011 ORA-17012 ORA-17014 ORA-17015 ORA-17016 ORA-17017 ORA-17018 ORA-17019 ORA-17020 ORA-17021 ORA-17022 ORA-17023 ORA-17024 ORA-17025 ORA-17026 ORA-17027 ORA-17028 ORA-17029 ORA-17030 ORA-17031 ORA-17032 ORA-17033 ORA-17034 ORA-17035 ORA-17036 ORA-17037 ORA-17038

Message Internal Error Io exception Invalid column index Invalid column type Unsupported column type Invalid column name Invalid dynamic column Closed Connection Closed Statement Closed Resultset Exhausted Resultset Parameter Type Conflict ResultSet.next was not called Statement was cancelled Statement timed out Cursor already initialized Invalid cursor Can only describe a query Invalid row prefetch Missing defines Missing defines at index Unsupported feature No data read Error in defines.isNull () Numeric Overflow Stream has already been closed Can not do new defines until the current ResultSet is closed setReadOnly: Read-only connections not supported READ_COMMITTED and SERIALIZABLE are the only valid transaction levels setAutoClose: Only support auto close mode on cannot set row prefetch to zero Malformed SQL92 string at position Non supported SQL92 token at position Character Set Not Supported !! exception in OracleNumber Fail to convert between UTF8 and UCS2 Byte array not long enough

C-2 Oracle Database JDBC Developer's Guide and Reference

General JDBC Messages

ORA Number ORA-17039 ORA-17040 ORA-17041 ORA-17042 ORA-17043 ORA-17044 ORA-17045 ORA-17046 ORA-17047 ORA-17048 ORA-17049 ORA-17050 ORA-17051 ORA-17052 ORA-17053 ORA-17054 ORA-17055 ORA-17056 ORA-17057 ORA-17058 ORA-17059 ORA-17060 ORA-17061 ORA-17062 ORA-17063 ORA-17064 ORA-17065 ORA-17066 ORA-17067 ORA-17068 ORA-17069 ORA-17070 ORA-17071 ORA-17072 ORA-17073 ORA-17074 ORA-17075

Message Char array not long enough Sub Protocol must be specified in connection URL Missing IN or OUT parameter at index: Invalid Batch Value Invalid stream maximum size Internal error: Data array not allocated Internal error: Attempt to access bind values beyond the batch value Internal error: Invalid index for data access Error in Type Descriptor parse Undefined type Inconsistent java and sql object types no such element in vector This API cannot be be used for non-UDT types This ref is not valid The size is not valid The LOB locator is not valid Invalid character encountered in Non supported character set (add orai18n.jar in your classpath environment variable) Closed LOB Internal error: Invalid NLS Conversion ratio Fail to convert to internal representation Fail to construct descriptor Missing descriptor Ref cursor is invalid Not in a transaction Invalid Sytnax or Database name is null Conversion class is null Access layer specific implementation needed Invalid Oracle URL specified Invalid argument(s) in call Use explicit XA call Data size bigger than max size for this type Exceeded maximum VARRAY limit Inserted value too large for column Logical handle no longer valid invalid name pattern Invalid operation for forward only resultset

JDBC Error Messages C-3

General JDBC Messages

ORA Number ORA-17076 ORA-17077 ORA-17078 ORA-17079 ORA-17080 ORA-17081 ORA-17082 ORA-17083 ORA-17084 ORA-17085 ORA-17086 ORA-17087 ORA-17088 ORA-17089 ORA-17090 ORA-17091 ORA-17092 ORA-17093 ORA-17094 ORA-17095 ORA-17096 ORA-17097 ORA-17098 ORA-17099 ORA-17100 ORA-17101 ORA-17102 ORA-17103 ORA-17104 ORA-17105 ORA-17106 ORA-17107 ORA-17108 ORA-17109 ORA-17110

Message Invalid operation for read only resultset Fail to set REF value Cannot do the operation as connections are already opened User credentials doesn't match the existing ones invalid batch command error occurred during batching No current row Not on the insert row Called on the insert row Value conflicts occurs Undefined column value on the insert row Ignored performance hint: setFetchDirection() Unsupported syntax for requested resultset type and concurrency level internal error operation not allowed Unable to create resultset at the requested type and/or concurrency level JDBC statements cannot be created or executed at end of call processing OCI operation returned OCI_SUCCESS_WITH_INFO Object type version mismatched Statement Cache size has not been set Statement Caching cannot be enabled for this logical connection Invalid PL/SQL Index Table element type Invalid empty lob operation Invalid PL/SQL Index Table array length Invalid database Java Object Invalid properties in OCI Connection Pool Object Bfile is read only invalid connection type to return via getConnection. Use getJavaSqlConnection instead SQL statement to execute cannot be empty or null connection session time zone was not set invalid combination of connections specified invalid proxy type specified No max length specified in defineColumnType standard Java character encoding not found execution completed with warning

C-4 Oracle Database JDBC Developer's Guide and Reference

General JDBC Messages

ORA Number ORA-17111 ORA-17112 ORA-17113 ORA-17114 ORA-17115 ORA-17116 ORA-17117 ORA-17118 ORA-17119 ORA-17120 ORA-17121 ORA-17122 ORA-17123 ORA-17124 ORA-17125 ORA-17126 ORA-17127

Message Invalid connection cache TTL timeout specified Invalid thread interval specified Thread interval value is more than the cache timeout value could not use local transaction commit in a global transaction could not use local transaction rollback in a global transaction could not turn on auto-commit in an active global transaction could not set savepoint in an active global transaction could not obtain ID for a named Savepoint could not obtain name for an un-named Savepoint could not set a Savepoint with auto-commit on could not rollback to a Savepoint with auto-commit on could not rollback to a local txn Savepoint in a global transaction Invalid statement cache size specified Invalid connection cache Inactivity timeout specified Improper statement type returned by explicit cache Fixed Wait timeout elapsed Invalid Fixed Wait timeout specified

JDBC Messages Sorted in Alphabetic Order

The following table lists the JDBC error messages sorted in alphabetic order:
ORA Number ORA-17066 ORA-17102 ORA-17038 ORA-17084 ORA-17028 ORA-17019 ORA-17078 ORA-17032 ORA-17039 ORA-17035 ORA-17008 ORA-17057 ORA-17010 ORA-17009 ORA-17105 ORA-17065 Message Access layer specific implementation needed Bfile is read only Byte array not long enough Called on the insert row Can not do new defines until the current ResultSet is closed Can only describe a query Cannot do the operation as connections are already opened cannot set row prefetch to zero Char array not long enough Character Set Not Supported !! Closed Connection Closed LOB Closed Resultset Closed Statement connection session time zone was not set Conversion class is null

JDBC Error Messages C-5

General JDBC Messages

ORA Number ORA-17118 ORA-17119 ORA-17122 ORA-17121 ORA-17120 ORA-17117 ORA-17116 ORA-17114 ORA-17115 ORA-17017 ORA-17070 ORA-17025 ORA-17047 ORA-17081 ORA-17071 ORA-17036 ORA-17110 ORA-17011 ORA-17060 ORA-17037 ORA-17059 ORA-17077 ORA-17126 ORA-17087 ORA-17125 ORA-17049 ORA-17072 ORA-17089 ORA-17001 ORA-17045 ORA-17044 ORA-17046 ORA-17058 ORA-17068 ORA-17080 ORA-17042 ORA-17055

Message could not obtain ID for a named Savepoint could not obtain name for an un-named Savepoint could not rollback to a local txn Savepoint in a global transaction could not rollback to a Savepoint with auto-commit on could not set a Savepoint with auto-commit on could not set savepoint in an active global transaction could not turn on auto-commit in an active global transaction could not use local transaction commit in a global transaction could not use local transaction rollback in a global transaction Cursor already initialized Data size bigger than max size for this type Error in defines.isNull () Error in Type Descriptor parse error occurred during batching Exceeded maximum VARRAY limit exception in OracleNumber execution completed with warning Exhausted Resultset Fail to construct descriptor Fail to convert between UTF8 and UCS2 Fail to convert to internal representation Fail to set REF value Fixed Wait timeout elapsed Ignored performance hint: setFetchDirection() Improper statement type returned by explicit cache Inconsistent java and sql object types Inserted value too large for column internal error Internal Error Internal error: Attempt to access bind values beyond the batch value Internal error: Data array not allocated Internal error: Invalid index for data access Internal error: Invalid NLS Conversion ratio Invalid argument(s) in call invalid batch command Invalid Batch Value Invalid character encountered in

C-6 Oracle Database JDBC Developer's Guide and Reference

General JDBC Messages

ORA Number ORA-17003 ORA-17006 ORA-17004 ORA-17106 ORA-17124 ORA-17111 ORA-17103 ORA-17018 ORA-17100 ORA-17007 ORA-17098 ORA-17127 ORA-17074 ORA-17075 ORA-17076 ORA-17067 ORA-17099 ORA-17097 ORA-17101 ORA-17107 ORA-17020 ORA-17123 ORA-17043 ORA-17064 ORA-17112 ORA-17002 ORA-17092 ORA-17073 ORA-17033 ORA-17021 ORA-17022 ORA-17061 ORA-17041 ORA-17082 ORA-17024 ORA-17108 ORA-17050

Message Invalid column index Invalid column name Invalid column type invalid combination of connections specified Invalid connection cache Inactivity timeout specified Invalid connection cache TTL timeout specified invalid connection type to return via getConnection. Use getJavaSqlConnection instead Invalid cursor Invalid database Java Object Invalid dynamic column Invalid empty lob operation Invalid Fixed Wait timeout specified invalid name pattern Invalid operation for forward only resultset Invalid operation for read only resultset Invalid Oracle URL specified Invalid PL/SQL Index Table array length Invalid PL/SQL Index Table element type Invalid properties in OCI Connection Pool Object invalid proxy type specified Invalid row prefetch Invalid statement cache size specified Invalid stream maximum size Invalid Sytnax or Database name is null Invalid thread interval specified Io exception JDBC statements cannot be created or executed at end of call processing Logical handle no longer valid Malformed SQL92 string at position Missing defines Missing defines at index Missing descriptor Missing IN or OUT parameter at index: No current row No data read No max length specified in defineColumnType no such element in vector

JDBC Error Messages C-7

General JDBC Messages

ORA Number ORA-17056 ORA-17034 ORA-17063 ORA-17083 ORA-17026 ORA-17094 ORA-17093 ORA-17090 ORA-17012 ORA-17030 ORA-17062 ORA-17014 ORA-17031 ORA-17029 ORA-17104 ORA-17109 ORA-17096 ORA-17095 ORA-17016 ORA-17015 ORA-17027 ORA-17040 ORA-17054 ORA-17053 ORA-17051 ORA-17052 ORA-17113 ORA-17091 ORA-17086 ORA-17048 ORA-17005 ORA-17023 ORA-17088 ORA-17069 ORA-17079 ORA-17085

Message Non supported character set Non supported SQL92 token at position Not in a transaction Not on the insert row Numeric Overflow Object type version mismatched OCI operation returned OCI_SUCCESS_WITH_INFO operation not allowed Parameter Type Conflict READ_COMMITTED and SERIALIZABLE are the only valid transaction levels Ref cursor is invalid ResultSet.next was not called setAutoClose: Only support auto close mode on setReadOnly: Read-only connections not supported SQL statement to execute cannot be empty or null standard Java character encoding not found Statement Caching cannot be enabled for this logical connection Statement Cache size has not been set Statement timed out Statement was cancelled Stream has already been closed Sub Protocol must be specified in connection URL The LOB locator is not valid The size is not valid This API cannot be be used for non-UDT types This ref is not valid Thread interval value is more than the cache timeout value Unable to create resultset at the requested type and/or concurrency level Undefined column value on the insert row Undefined type Unsupported column type Unsupported feature Unsupported syntax for requested resultset type and concurrency level Use explicit XA call User credentials doesn't match the existing ones Value conflicts occurs

C-8 Oracle Database JDBC Developer's Guide and Reference

Native XA Messages

Native XA Messages
The following sections cover the JDBC error messages that are specific to the Native XA feature:

Native XA Messages Sorted by ORA Number Native XA Messages Sorted in Alphabetic Order

Native XA Messages Sorted by ORA Number

The following table lists the Native XA messages sorted by the ORA number:
ORA Number ORA-17200 ORA-17201 ORA-17202 ORA-17203 ORA-17204 ORA-17205 ORA-17206 ORA-17207 ORA-17213 ORA-17215 ORA-17216 ORA-17233 ORA-17235 ORA-17236 Message Unable to properly convert XA open string from Java to C Unable to properly convert XA close string from Java to C Unable to properly convert RM name from Java to C Could not cast pointer type to jlong Input array too short to hold OCI handles Failed to obtain OCISvcCtx handle from C-XA using xaoSvcCtx Failed to obtain OCIEnv handle from C-XA using xaoEnv The tnsEntry property was not set in DataSource C-XA returned XAER_RMERR during xa_open C-XA returned XAER_INVAL during xa_open C-XA returned XAER_PROTO during xa_open C-XA returned XAER_RMERR during xa_close C-XA returned XAER_INVAL during xa_close C-XA returned XAER_PROTO during xa_close

Native XA Messages Sorted in Alphabetic Order

The following table lists the Native XA messages sorted in the alphabetic order:
ORA Number ORA-17203 ORA-17235 ORA-17215 ORA-17236 ORA-17216 ORA-17233 ORA-17213 ORA-17206 ORA-17205 ORA-17204 ORA-17207 Message Could not cast pointer type to jlong C-XA returned XAER_INVAL during xa_close C-XA returned XAER_INVAL during xa_open C-XA returned XAER_PROTO during xa_close C-XA returned XAER_PROTO during xa_open C-XA returned XAER_RMERR during xa_close C-XA returned XAER_RMERR during xa_open Failed to obtain OCIEnv handle from C-XA using xaoEnv Failed to obtain OCISvcCtx handle from C-XA using xaoSvcCtx Input array too short to hold OCI handles The tnsEntry property was not set in DataSource

JDBC Error Messages C-9

TTC Messages

ORA Number ORA-17202 ORA-17201 ORA-17200

Message Unable to properly convert RM name from Java to C Unable to properly convert XA close string from Java to C Unable to properly convert XA open string from Java to C

TTC Messages
This section lists TTC error messages, first sorted by the ORA number and then in alphabetic order in the following subsections:

TTC Messages Sorted by ORA Number TTC Messages Sorted in Alphabetic Order

TTC Messages Sorted by ORA Number

The following table lists the TTC messages sorted by the ORA number:
ORA Number ORA-17401 ORA-17402 ORA-17403 ORA-17404 ORA-17405 ORA-17406 ORA-17407 ORA-17408 ORA-17409 ORA-17410 ORA-17411 ORA-17412 ORA-17413 ORA-17414 ORA-17415 ORA-17416 ORA-17417 ORA-17418 ORA-17419 ORA-17420 ORA-17421 ORA-17422 ORA-17423 ORA-17424 Message Protocol violation Only one RPA message is expected Only one RXH message is expected Received more RXDs than expected UAC length is not zero Exceeding maximum buffer length invalid Type Representation(setRep) invalid Type Representation(getRep) invalid buffer length No more data to read from socket Data Type representations mismatch Bigger type length than Maximum Exceding key size Insufficient Buffer size to store Columns Names This type hasn't been handled FATAL NLS Problem, failed to decode column names Internal structure's field length error Invalid number of columns returned Oracle Version not defined Types or Connection not defined Invalid class in factory Using a PLSQL block without an IOV defined Attempting different marshaling operation

C-10 Oracle Database JDBC Developer's Guide and Reference

TTC Messages

ORA Number ORA-17425 ORA-17426 ORA-17427 ORA-17428 ORA-17429 ORA-17430 ORA-17431 ORA-17432 ORA-17433 ORA-17434 ORA-17435 ORA-17436 ORA-17437 ORA-17438 ORA-17439 ORA-17440 ORA-17441 ORA-17442 ORA-17443 ORA-17444

Message Returning a stream in PLSQL block Both IN and OUT binds are NULL Using Uninitialized OAC Logon must be called after connect Must be at least connected to server Must be logged on to server SQL Statement to parse is null invalid options in all7 invalid arguments in call not in streaming mode invalid number of in_out_binds in IOV invalid number of outbinds Error in PLSQL block IN/OUT argument(s) Internal - Unexpected value Invalid SQL type DBItem/DBType is null Oracle Version not supported. Minimum supported version is 7.2.3. Refcursor value is invalid Null user or password not supported in THIN driver TTC Protocol version received from server not supported

TTC Messages Sorted in Alphabetic Order

The following table lists the TTC messages in the alphabetic order:
ORA Number ORA-17424 ORA-17412 ORA-17426 ORA-17411 ORA-17440 ORA-17437 ORA-17413 ORA-17406 ORA-17416 ORA-17414 ORA-17438 ORA-17418 ORA-17433 Message Attempting different marshaling operation Bigger type length than Maximum Both IN and OUT binds are NULL Data Type representations mismatch DBItem/DBType is null Error in PLSQL block IN/OUT argument(s) Exceding key size Exceeding maximum buffer length FATAL Insufficient Buffer size to store Columns Names Internal - Unexpected value Internal structure's field length error invalid arguments in call

JDBC Error Messages C-11

TTC Messages

ORA Number ORA-17409 ORA-17422 ORA-17419 ORA-17435 ORA-17436 ORA-17432 ORA-17439 ORA-17408 ORA-17407 ORA-17428 ORA-17429 ORA-17430 ORA-17417 ORA-17410 ORA-17434 ORA-17443 ORA-17402 ORA-17403 ORA-17420 ORA-17441 ORA-17401 ORA-17404 ORA-17442 ORA-17425 ORA-17431 ORA-17415 ORA-17444 ORA-17421 ORA-17405 ORA-17423 ORA-17427

Message invalid buffer length Invalid class in factory Invalid number of columns returned invalid number of in_out_binds in IOV invalid number of outbinds invalid options in all7 Invalid SQL type invalid Type Representation(getRep) invalid Type Representation(setRep) Logon must be called after connect Must be at least connected to server Must be logged on to server NLS Problem, failed to decode column names No more data to read from socket not in streaming mode Null user or password not supported in THIN driver Only one RPA message is expected Only one RXH message is expected Oracle Version not defined Oracle Version not supported. Minimum supported version is 7.2.3. Protocol violation Received more RXDs than expected Refcursor value is invalid Returning a stream in PLSQL block SQL Statement to parse is null This type hasn't been handled TTC Protocol version received from server not supported Types or Connection not defined UAC length is not zero Using a PLSQL block without an IOV defined Using Uninitialized OAC

C-12 Oracle Database JDBC Developer's Guide and Reference

D
Troubleshooting
This appendix describes how to troubleshoot a Java Database Connectivity (JDBC) application or applet, and contains the following topics:

Common Problems Basic Debugging Procedures

Common Problems
This section describes some common problems that you might encounter while using Oracle JDBC drivers. These problems include:

Memory Consumption for CHAR Columns Defined as OUT or IN/OUT Variables Memory Leaks and Running Out of Cursors Boolean Parameters in PL/SQL Stored Procedures Opening More Than 16 OCI Connections for a Process Using statement.cancel Using JDBC with Firewalls

Memory Consumption for CHAR Columns Defined as OUT or IN/OUT Variables

In PL/SQL, when a CHAR or a VARCHAR2 column is defined as a OUT or IN/OUT variable, the driver allocates a CHAR array of 32512 chars. This can cause a memory consumption problem. JDBC Thin driver does not allocate memory when using VARCHAR2 output type. But JDBC OCI driver allocates memory for both CHAR and VARCHAR2 types. So, CPU load in OCI driver is higher than Thin driver. At previous releases, the solution to the problem was to invoke the Statement.setMaxFieldSize method. A better solution is to use OracleCallableStatement.registerOutParameter. Oracle encourages you always to call registerOutParameter (int paramIndex, int sqlType, int scale, int maxLength) on each CHAR or VARCHAR2 column. This method is defined in oracle.jdbc.driver.OracleCallableStatement. Use the fourth argument, maxLength, to limit the memory consumption. This parameter tells the driver how many characters are necessary to store this column. The column is truncated if the character array cannot hold the column data. The third argument, scale, is ignored by the driver.

Troubleshooting D-1

Common Problems

Memory Leaks and Running Out of Cursors

If you receive messages that you are running out of cursors or that you are running out of memory, make sure that all your Statement and ResultSet objects are explicitly closed. Oracle JDBC drivers do not have finalizer methods. They perform cleanup routines by using the close method of the ResultSet and Statement classes. If you do not explicitly close your result set and statement objects, significant memory leaks can occur. You could also run out of cursors in the database. Closing a statement releases the corresponding cursor in the database. Similarly, you must explicitly close Connection objects to avoid leaking and running out of cursors on the server-side. When you close the connection, the JDBC driver closes any open statement objects associated with it, thus releasing the cursor on the server-side.

Boolean Parameters in PL/SQL Stored Procedures

The JDBC drivers do not support the passing of BOOLEAN parameters to PL/SQL stored procedures. If a PL/SQL procedure contains BOOLEAN values, you can work around the restriction by wrapping the PL/SQL procedure with a second PL/SQL procedure that accepts the argument as an INT and passes it to the first stored procedure. When the second procedure is called, the server performs the conversion from INT to BOOLEAN. The following is an example of a stored procedure, BOOLPROC, that attempts to pass a BOOLEAN parameter, and a second procedure, BOOLWRAP, that performs the substitution of an INT value for the BOOLEAN.
CREATE OR REPLACE PROCEDURE boolproc(x boolean) AS BEGIN [...] END; CREATE OR REPLACE PROCEDURE boolwrap(x int) AS BEGIN IF (x=1) THEN boolproc(TRUE); ELSE boolproc(FALSE); END IF; END; // Create the database connection from a DataSource OracleDataSource ods = new OracleDataSource(); ods.setURL("jdbc:oracle:oci:@<...hoststring...>"); ods.setUser("scott"); ods.setPassword("tiger"); Connection conn = ods.getConnection(); CallableStatement cs = conn.prepareCall ("begin boolwrap(?); end;"); cs.setInt(1, 1); cs.execute ();

Opening More Than 16 OCI Connections for a Process

You might find that you are not able to open more than approximately 16 JDBC-OCI connections for a process at any given time. The most likely reasons for this would be either that the number of processes on the server exceeded the limit specified in the
D-2 Oracle Database JDBC Developer's Guide and Reference

Common Problems

initialization file, or that the per-process file descriptors limit was exceeded. It is important to note that one JDBC-OCI connection can use more than one file descriptor (it might use anywhere between 3 and 4 file descriptors). If the server allows more than 16 processes, then the problem could be with the per-process file descriptor limit. The possible solution would be to increase this limit.

Using statement.cancel
The JDBC standard method Statement.cancel attempts to cleanly stop the execution of a SQL statement by sending a message to the database. In response, the database stops execution and replies with an error message. The Java thread that invoked Statement.execute waits on the server, and continues execution only when it receives the error reply message invoked by the other thread's call to Statement.cancel. As a result, Statement.cancel relies on the correct functioning of the network and the database. If either the network connection is broken or the database server is hung, the client does not receive the error reply to the cancel message. Frequently, when the server process dies, JDBC receives an IOException that frees the thread that invoked Statement.execute. In some circumstances, the server is hung, but JDBC does not receive an IOException. Statement.cancel does not free the thread that initiated the Statement.execute. When JDBC does not receive an IOException, Oracle Net may eventually time out and close the connection. This causes an IOException and frees the thread. This process can take many minutes. For information about how to control this time-out, see the description of the readTimeout property for OracleDatasource.setConnectionProperties. You can also tune this time-out with certain Oracle Net settings. See the Oracle Database Net Services Administrator's Guide for more information. The JDBC standard method Statement.setQueryTimeout relies on Statement.cancel. If execution continues longer than the specified time-out interval, then the monitor thread calls Statement.cancel. This is subject to all the same limitations described previously. As a result, there are cases when the time-out does not free the thread that invoked Statement.execute. The length of time between execution and cancellation is not precise. This interval is no less than the specified time-out interval but can be several seconds longer. If the application has active threads running at high priority, then the interval can be arbitrarily longer. The monitor thread runs at high priority, but other high priority threads may keep it from running indefinitely. Note that the monitor thread is started only if there are statements executed with non zero time-out. There is only one monitor thread that monitors all Oracle JDBC statement execution. Statement.cancel and Statement.setQueryTimeout are not supported in the server-side internal driver. The server-side internal driver runs in the single-threaded server process; the Oracle JVM implements Java threads within this single-threaded process. If the server-side internal driver is executing a SQL statement, then no Java thread can call Statement.cancel. This also applies to the Oracle JDBC monitor thread.

Using JDBC with Firewalls

Firewall timeout for idle-connections may sever a connection. This can cause JDBC applications to hang while waiting for a connection. You can perform one or more of the following actions to avoid connections from being severed due to firewall timeout:

Troubleshooting D-3

Basic Debugging Procedures

If you are using connection caching or connection pooling, then always set the inactivity timeout value on the connection cache to be shorter than the firewall idle timeout value. Pass oracle.net.READ_TIMEOUT as connection property to enable read timeout on socket. The timeout value is in milliseconds. For both JDBC OCI and JDBC Thin drivers, use net descriptor to connect to the database and specify the ENABLE=BROKEN parameter in the DESCRIPTION clause in the connect descriptor. Also, set a lower value for tcp_keepalive_interval. Enable Oracle Net DCD by setting SQLNET.EXPIRE_TIME=1 in the sqlnet.ora file on the server-side.

Basic Debugging Procedures

This section describes strategies for debugging a JDBC program:

Oracle Net Tracing to Trap Network Events Third Party Debugging Tools

For information about processing SQL exceptions, including printing stack traces to aid in debugging, see "Processing SQL Exceptions" on page 2-14.

Oracle Net Tracing to Trap Network Events

You can enable client and server Oracle-Net trace to trap the packets sent over Oracle Net. You can use client-side tracing only for the JDBC OCI driver; it is not supported for the JDBC Thin driver. You can find more information about tracing and reading trace files in the Oracle Net Services Administrator's Guide. The trace facility produces a detailed sequence of statements that describe network events as they execute. "Tracing" an operation lets you obtain more information about the internal operations of the event. This information is printed to a readable file that identifies the events that led to the error. Several Oracle Net parameters in the SQLNET.ORA file control the gathering of trace information. After setting the parameters in SQLNET.ORA, you must make a new connection for tracing to be performed. The higher the trace level, the more detail is captured in the trace file. Because the trace file can be hard to understand, start with a trace level of 4 when enabling tracing. The first part of the trace file contains connection handshake information, so look beyond this for the SQL statements and error messages related to your JDBC program.
Note:

The trace facility uses a large amount of disk space and might have significant impact upon system performance. Therefore, enable tracing only when necessary.

Client-Side Tracing
Set the following parameters in the SQLNET.ORA file on the client system. TRACE_LEVEL_CLIENT Purpose: Turns tracing on/off to a certain specified level. Default Value:
D-4 Oracle Database JDBC Developer's Guide and Reference

Basic Debugging Procedures

0 or OFF Available Values:

0 or OFF - No trace output 4 or USER - User trace information 10 or ADMIN - Administration trace information 16 or SUPPORT - WorldWide Customer Support trace information

Example: TRACE_LEVEL_CLIENT=10 TRACE_DIRECTORY_CLIENT Purpose: Specifies the destination directory of the trace file. Default Value: ORACLE_HOME/network/trace Example: UNIX: TRACE_DIRECTORY_CLIENT=/oracle/traces Windows: TRACE_DIRECTORY_CLIENT=C:\ORACLE\TRACES TRACE_FILE_CLIENT Purpose: Specifies the name of the client trace file. Default Value: SQLNET.TRC Example: TRACE_FILE_CLIENT=cli_Connection1.trc
Note:

Ensure that the name you choose for the TRACE_FILE_CLIENT file is different from the name you choose for the TRACE_FILE_SERVER file.

TRACE_UNIQUE_CLIENT Purpose: Gives each client-side trace a unique name to prevent each trace file from being overwritten with the next occurrence of a client trace. The PID is attached to the end of the file name. Default Value: OFF Example: TRACE_UNIQUE_CLIENT = ON

Troubleshooting D-5

Basic Debugging Procedures

Server-Side Tracing
Set the following parameters in the SQLNET.ORA file on the server system. Each connection will generate a separate file with a unique file name. TRACE_LEVEL_SERVER Purpose: Turns tracing on/off to a certain specified level. Default Value: 0 or OFF Available Values:

0 or OFF - No trace output 4 or USER - User trace information 10 or ADMIN - Administration trace information 16 or SUPPORT - WorldWide Customer Support trace information

Example: TRACE_LEVEL_SERVER=10 TRACE_DIRECTORY_SERVER Purpose: Specifies the destination directory of the trace file. Default Value: ORACLE_HOME/network/trace Example: TRACE_DIRECTORY_SERVER=/oracle/traces TRACE_FILE_SERVER Purpose: Specifies the name of the server trace file. Default Value: SERVER.TRC Example: TRACE_FILE_SERVER= svr_Connection1.trc
Note:

Ensure that the name you choose for the TRACE_FILE_SERVER file is different from the name you choose for the TRACE_FILE_CLIENT file.

Third Party Debugging Tools

You can use tools such as JDBCSpy and JDBCTest from Intersolv to troubleshoot at the JDBC API level. These tools are similar to ODBCSpy and ODBCTest.

D-6 Oracle Database JDBC Developer's Guide and Reference

Index
A
absolute positioning in result sets, 17-2 acceptChanges() method, 18-8 Accessing PL/SQL Index-by Tables, 4-4 addBatch() method, 23-9 APPLET HTML tag, 5-8 applets connecting to a database, 5-2 deploying in an HTML page, 5-8 packaging, 5-8 signed applets browser security, 5-5 object-signing certificate, 5-6 using signed applets, 5-5 using with firewalls, 5-6 ARCHIVE, parameter for APPLET tag, 5-9 ARRAY class, 4-8 descriptors, 4-8 objects, creating, 4-8, 16-5 arrays defined, 16-1 getting, 16-9 named, 16-1 passing to callable statement, 16-12 retrieving from a result set, 16-6 retrieving partial arrays, 16-8 using type maps, 16-12 working with, 16-1 authentication (security), 9-3 auto-commit, 2-11 auto-commit mode disabling, B-1 result set behavior, B-2 statement batch value, setting, 23-4 BatchUpdateException, 23-12 beforeFirst() method, 18-5 BFILE accessing data, 14-18 class, 4-8 creating and populating columns, 14-17 defined, 12-6 introduction, 14-1 locators, 14-15 getting from a result set, 14-15 getting from callable statement, 14-15 passing to callable statements, 14-16 passing to prepared statements, 14-16 manipulating data, 14-18 reading data, 14-16 BFILE locator, selecting, 4-9 BigDecimal mapping (for attributes), 13-32 BLOB, 14-4 class, 4-8 creating and populating, 14-8 getting locators, 14-2 introduction, 14-1 locators getting from result set, 14-3 selecting, 4-9 manipulating data, 14-10 reading data, 14-5, 14-6 writing data, 14-7 BLOBs size limit with PL/SQL prodcedures, 14-4 Boolean parameters, restrictions, D-2 branch qualifier (distributed transactions), 28-12

C
CachedRowSet, 18-6 caching, client-side custom use for scrollable result sets, 17-4 Oracle use for scrollable result sets, 17-3 callable statement getting a BFILE locator, 14-15 getting LOB locators, 14-3 passing BFILE locator, 14-16 passing LOB locators, 14-4 using getOracleObject() method, 11-7

B
batch jobs, authenticating users in, 9-21 batch updates--see update batching batch value checking value, 23-5 connection batch value, setting, 23-4 connection vs. statement value, 23-3 default value, 23-3 overriding value, 23-5

Index-1

cancel() method, 4-21 cancelling SQL statements, D-3 cancelRowUpdates() method (result set), 17-12 casting return values, 11-10 catalog arguments (DatabaseMetaData), A-12 CHAR columns using setFixedCHAR() to match in WHERE, 11-13 character sets, 4-13 checksums code example, 9-9 setting parameters in Java, 9-9 support by OCI drivers, 9-7 support by Thin driver, 9-8 CLASSPATH environment variable, specifying, 2-3 clearBatch() method, 23-10 clearDefines() method, 23-17 CLOB class, 4-8 creating and populating, 14-8 introduction, 14-1 locators, 14-2 getting from result set, 14-3 passing to callable statements, 14-4 passing to prepared statement, 14-4 locators, selecting, 4-9 manipulating data, 14-10 reading data, 14-5, 14-6 writing data, 14-7, 14-8 close method, 20-9 close(), 20-3 close() method, D-2 for caching statements, 20-5, 20-6 closeWithKey(), 20-3 closeWithKey() method, 20-7 CMAN.ORA file, creating, 5-4 CODE, parameter for APPLET tag, 5-8 CODEBASE, parameter for APPLET tag, 5-9 collections defined, 16-1 collections (nested tables and arrays), 16-5 column types defining, 23-17 redefining, 23-15 commit a distributed transaction branch, 28-11 commit changes to database, 2-11 commit() method, 4-21 CONCUR_READ_ONLY result sets, 17-5 CONCUR_UPDATABLE result sets, 17-5 concurrency types in result sets, 17-3 CONNECT / feature, 9-21 connection closing, 2-12 opening, 2-8 connection attributes, 21-6 connection cache properties, 21-8 Connection Manager installing, 5-4 starting, 5-4

using, 5-4 using multiple managers, 5-5 writing the connection string, 5-5 connection properties, 8-8 put() method, 8-10 connection string Connection Manager, 5-5 connections read-only, B-4 constants for SQL types, 4-23 CREATE DIRECTORY statement for BFILEs, 14-17 CREATE TABLE statement to create BFILE columns, 14-17 to create BLOB, CLOB columns, 14-9 CREATE TYPE statement, 13-20 create() method for ORADataFactory interface, 13-15 createDescriptor() method, 13-5 createStatement(), 20-3 createStatement() method, 20-7 createTemporary() method, 14-13 CursorName limitations, A-11 cursors, D-2 custom collection classes and JPublisher, 16-14 defined, 16-1, 16-13 custom Java classes, 4-3 defined, 13-1 custom object classes creating, 13-6 defined, 13-1 custom reference classes and JPublisher, 15-5 defined, 15-1, 15-5

D
data conversions, 11-3 LONG, 12-2 LONG RAW, 12-2 data sources creating and connecting (with JNDI), 8-6 creating and connecting (without JNDI), 8-6 Oracle implementation, 8-2 properties, 8-2 standard interface, 8-2 data streaming avoiding, 12-5 data type classes, 4-5 data type mappings, 11-1 data types Java, 11-1 Java native, 11-1 JDBC, 11-1 Oracle SQL, 11-1 database connecting from an applet, 5-2

Index-2

via multiple Connection Managers, 5-5 with server-side internal driver, 7-1 connection testing, 2-5 database specifiers, 8-11 database URL including userid and password, 2-9 database URL, specifying, 2-8 database URLs and database specifiers, 8-11 DatabaseMetaData calls, A-12 DatabaseMetaData class, A-9 datasources, 8-1 and JNDI, 8-6 to 8-7 DATE class, 4-9 DBMS_SERVICE.SERVICE_TIME, 22-2 DBMS_SERVICE.THROUGHPUT, 22-2 debugging JDBC programs, D-4 DEFAULT_CHARSET character set value, 4-12 defaultConnection() method, 7-1 defineColumnType() method, 4-22, 12-5, 23-17 DELETE in a result set, 17-11 deleteRow() method (result set), 17-11 deletesAreDetected() method (database metadata), 17-18 distributed transaction ID component, 28-12 distributed transactions branch qualifier, 28-12 check for same resource manager, 28-12 commit a transaction branch, 28-11 components and scenarios, 28-2 concepts, 28-2 distributed transaction ID component, 28-12 end a transaction branch, 28-10 example of implementation, 28-15 global transaction identifier, 28-12 ID format identifier, 28-12 introduction, 28-1 Oracle XA connection implementation, 28-6 Oracle XA data source implementation, 28-6 Oracle XA ID implementation, 28-12 Oracle XA optimizations, 28-15 Oracle XA resource implementation, 28-7 prepare a transaction branch, 28-11 roll back a transaction branch, 28-11 start a transaction branch, 28-8 transaction branch ID component, 28-12 XA connection interface, 28-6 XA data source interface, 28-6 XA error handling, 28-14 XA exception classes, 28-13 XA ID interface, 28-12 XA resource functionality, 28-8 XA resource interface, 28-7 DML Returning, 4-4, 4-26 example, 4-27 limitations, 4-28 Oracle-specific APIs, 4-26 running statements, 4-27 Double.NaN restrictions on use, 4-9

driverType, 8-4

E
encryption code example, 9-9 overview, 9-6 setting parameters in Java, 9-9 support by OCI drivers, 9-7 support by Thin driver, 9-8 end a distributed transaction branch, 28-10 Enterprise Java Beans (EJB), 18-8 environment variables specifying, 2-3 errors general JDBC message structure, C-1 general JDBC messages, listed, C-1 processing exceptions, 2-14 TTC messages, listed, C-10 exceptions retrieving error code, 2-15 retrieving message, 2-15 retrieving SQL state, 2-15 execute() method, 18-10 executeBatch() method, 23-10 executeUpdate() method, 23-6 explicit statement caching definition of, 20-3 null data, 20-8 extensions to JDBC, Oracle, 4-1, 11-1, 13-1, 15-1, 16-1, 23-1 external changes (result set) defined, 17-16 seeing, 17-17 visibility vs. detection, 17-18 external file defined, 12-6

F
failover fast connection, 26-1 to 26-8 Fast Connection Failover, 26-1 to 26-8 fast connection failover prerequisites, 26-2 fetch direction in result sets, 17-10 fetch size, result sets, 17-15 FilteredRowSet, 18-12 finalizer methods, D-2 firewalls configuring for applets, 5-6 connection string, 5-7 described, 5-6 required rule list items, 5-6 using with applets, 5-6 Firewalls, using with JDBC, D-3 floating-point compliance, A-12 Float.NaN restrictions on use, 4-9 format identifier, transaction ID, 28-12

Index-3

forward-only result sets, 17-2 freeTemporary() method, 14-13 function call syntax, SQL92 syntax, A-10

G
getARRAY() method, 16-6 getArray() method, 16-4, 16-6 using type maps, 16-8 getAsciiStream() method for reading CLOB data, 14-5 getAttributes() method used by Structs, 13-10 getAutoBuffering() method of the oracle.sql.ARRAY class, 16-4 of the oracle.sql.STRUCT class, 13-6 getBaseType() method, 16-9 getBaseTypeName() method, 15-2 getBinaryStream() method, 12-3 for reading BFILE data, 14-16 for reading BLOB data, 14-5 getBytes() method, 4-6, 12-4 getCallWithKey(), 20-3 getCallWithKey() method, 20-7, 20-8 getCharacterStream() method for reading CLOB data, 14-5 getColumns() method, 23-19 getConcurrency() method (result set), 17-7 getConnection() method, 7-1, 24-6 getCursor() method, 4-15, 4-16 getCursorName() method limitations, A-11 getDefaultExecuteBatch() method, 4-21, 23-5 getDefaultRowPrefetch() method, 4-21 getDescriptor() method, 13-3 getDirAlias() method, 14-18 getErrorCode() method (SQLException), 2-15 getExecuteBatch() method, 23-4, 23-5 getFetchSize() method, 17-15 getJavaSQLConnection() method, 13-3 getJavaSqlConnection() method, 4-25 getMessage() method (SQLException), 2-15 getName() method, 14-18 getNumericFunctions() method, A-9 getObject() method casting return values, 11-10 for object references, 15-3 for ORAData objects, 13-15 for SQLInput streams, 13-11 for SQLOutput streams, 13-12 for Struct objects, 13-4 return types, 11-6, 11-7 to get BFILE locators, 14-15 to get Oracle objects, 13-4 used with ORAData interface, 13-17 getOracleArray() method, 16-6, 16-9 getOracleAttributes() method, 13-3, 13-4 getOracleObject() method casting return values, 11-10 return types, 11-6, 11-7

using in callable statement, 11-7 using in result set, 11-6 getOraclePlsqlIndexTable() method, 4-29, 4-32 argument int paramIndex, 4-33 code example, 4-33 getORAData() method, 13-15, 13-16 getPassword() method, 8-3 getPlsqlIndexTable() method, 4-29, 4-32, 4-33 arguments Class primitiveType, 4-34 int paramIndex, 4-33 code example, 4-32, 4-34 getProcedureColumns() method, 23-19 getProcedures() method, 23-19 getREF() method, 15-4 getRow() method (result set), 17-9 getRowPrefetch() method, 4-22 getSQLState() method (SQLException), 2-15 getSQLTypeName() method, 13-3, 16-9 getStatementCacheSize() method code example, 20-5 getStatementWithKey(), 20-3 getStatementWithKey() method, 20-7, 20-8 getString() method, 4-13 to get ROWIDs, 4-14 getStringFunctions() method, A-9 getStringWithReplacement() method, 4-13 getSTRUCT() method, 13-4 getSubString() method for reading CLOB data, 14-5, 14-6 getSystemFunctions() method, A-9 getTimeDateFunctions() method, A-9 getTransactionIsolation() method, B-4 getType() method (result set), 17-7 getTypeMap() method, 13-9 getUpdateCounts() method (BatchUpdateException), 23-13 getValue() method, 15-3 for object references, 15-3 getXXX() methods casting return values, 11-10 for specific data types, 11-9 Oracle extended properties, 8-5 global transaction identifier (distributed transactions), 28-12 global transactions, 28-1 globalization, 19-1 to ?? using, 19-1

H
HEIGHT, parameter for APPLET tag, 5-8 HTML tags, to deploy applets, 5-8

I
IEEE 754 floating-point compliance, A-12 implicit connection cache, 21-1 example, 21-5

Index-4

implicit statement caching definition of, 20-2 Least Recently Used (LRU) scheme, 20-2 IN OUT parameter mode, 4-31 IN parameter mode, 4-29 INSERT in a result set, 17-13 INSERT INTO statement for creating BFILE columns, 14-17 insertRow() method (result set), 17-14 insertsAreDetected() method (database metadata), 17-18 installation directories and files, 2-2 verifying on the client, 2-2 Instant Client feature, 6-2 integrity code example, 9-9 overview, 9-6 setting parameters in Java, 9-9 support by OCI drivers, 9-7 support by Thin driver, 9-8 internal changes (result set) defined, 17-16 seeing, 17-17 isAfterLast() method (result set), 17-9 isBeforeFirst() method (result set), 17-9 isFirst() method (result set), 17-9 isLast() method (result set), 17-9 isSameRM() (distributed transactions), 28-12 isTemporary() method, 14-13

J
Java compiling and running, 2-4 data types, 11-1 native data types, 11-1 stored procedures, 2-14 stream data, 12-1 Java Naming and Directory Interface (JNDI), 8-1 Java Sockets, 1-3 Java Virtual Machine (JVM), 7-1 java.math, Java math packages, 2-8 java.sql, JDBC packages, 2-8 java.sql.Connection interface close method, 20-9 java.sql.SQLException() method, 2-14 java.sql.Statement interface close method, 20-9 java.sql.Struct class getSQLTypeName() method, 13-3 java.sql.Types class, 23-17 java.util.Map class, 16-9 java.util.Properties, 24-5 JDBC and IDEs, 1-5 basic program, 2-7 data types, 11-1 defined, 1-1 importing packages, 2-8

limitations of Oracle extensions, A-11 sample files, 2-4 testing, 2-5 version support, 3-1 to 3-6 JDBC 2.0 support data type support, 3-2 extended feature support, 3-2 introduction, 3-1 JDK 1.2.x vs. JDK 1.1.x, 3-1, 3-2 standard feature support, 3-2 JDBC drivers applets, 5-2 choosing a driver for your needs, 1-4 common features, 1-2 common problems, D-1 determining driver version, 2-5 introduction, 1-1 restrictions, D-2 SQL92 syntax, A-6 JDBC mapping (for attributes), 13-31 JdbcCheckup program, 2-5 JDBCRowSet, 18-9 JDBCSpy, D-6 JDBCTest, D-6 JDeveloper, 1-5 JDK versions supported, 1-5 JNDI and datasources, 8-6 to 8-7 looking up data source, 8-7 overview of Oracle support, 8-1 registering data source, 8-7 JoinRowSet, 18-13 JPublisher, 13-17, 13-30 JPublisher utility, 13-7 creating custom collection classes, 16-13 creating custom Java classes, 13-30 creating custom reference classes, 15-5 SQL type categories and mapping options, 13-31 type mapping modes and settings, 13-31 type mappings, 13-31 JVM, 7-1

K
KPRB driver overview, 1-3 relation to the SQL engine, 7-1 session context, 7-4 testing, 7-4 transaction context, 7-4 URL for, 7-2

L
LD_LIBRARY_PATH environment variable, specifying, 2-4 LDAP and SSL, 8-13 Least Recently Used (LRU) scheme, 20-2, 24-5

Index-5

libheteroxa11_g.so shared library, 28-20 libheteroxa11.so shared library, 28-20 LIKE escape characters, SQL92 syntax, A-9 limitations on setBytes() and setString(), use of streams to avoid, 12-9 Load Balancing Advisory, 22-2 LOB defined, 12-6 introduction, 14-1 locators, 14-2 reading data, 14-5 LOB locators getting from callable statements, 14-3 passing, 14-4 locators getting for BFILEs, 14-15 getting for BLOBs, 14-2 getting for CLOBs, 14-2 LOB, 14-2 passing to callable statements, 14-4 passing to prepared statement, 14-4 LONG data conversions, 12-2 LONG RAW data conversions, 12-2 LRU scheme, 20-2, 24-5

O
object references accessing object values, 15-3, 15-5 described, 15-1 passing to prepared statements, 15-4 retrieving, 15-3 retrieving from callable statement, 15-4 updating object values, 15-3, 15-5 object-JDBC mapping (for attributes), 13-31 OCI driver described, 1-3 ODBCSpy, D-6 ODBCTest, D-6 ONS configuring, 26-2 to 26-4 ons.config file, 26-2, 26-3, 26-4 optimization, performance, B-1 Oracle Advanced Security support by JDBC, 9-1 support by OCI drivers, 9-2 support by Thin driver, 9-3 Oracle Connection Manager, 5-3 Oracle data types using, 11-1 Oracle extensions, 4-2 data type support, 4-2 limitations, A-11 catalog arguments to DatabaseMetaData calls, A-12 CursorName, A-11 IEEE 754 floating-point compliance, A-12 PL/SQL TABLE, BOOLEAN, RECORD types, A-11 read-only connection, B-4 SQL92 outer join escapes, A-11 SQLWarning class, A-12 object support, 4-3 result sets, 11-5 statements, 11-5 to JDBC, 4-1, 11-1, 13-1, 15-1, 16-1, 23-1 Oracle JPublisher, 4-3 generated classes, 13-25 Oracle mapping (for attributes), 13-31 Oracle Notification Service. See ONS Oracle objects and JDBC, 13-1 converting with ORAData interface, 13-14 converting with SQLData interface, 13-10 getting with getObject() method, 13-4 Java classes which support, 13-2 mapping to custom object classes, 13-6 reading data by using SQLData interface, 13-12 working with, 13-1 writing data by using SQLData interface, 13-14 Oracle SQL data types, 11-1 OracleCallableStatement interface, 4-22 getOraclePlsqlIndexTable() method, 4-29 getPlsqlIndexTable() method, 4-29 getXXX() methods, 11-9 registerIndexTableOutParameter() method, 4-29,

M
make() method, 4-12 memory leaks, D-2 moveToCurrentRow() method (result set), 17-13 moveToInsertRow() method (result set), 17-13 mutable arrays, 16-13

N
named arrays, 16-1 defined, 16-5 nativeXA, 8-4, 28-20 NCLOB creating and populating, 14-8 reading data, 14-7 network events, trapping, D-4 next() method, 18-5 next() method (result set), 17-10 NLS. See globalization NLS_LANG variable desupported, 19-1 NULL testing for, 11-4 NULL data converting, 11-4 null data explicit statement caching, 20-8 NullPointerException thrown when converting Double.NaN and Float.NaN, 4-9 NUMBER class, 4-9

Index-6

4-31 registerOutParameter() method, 11-11 setPlsqlIndexTable() method, 4-29 OracleCallableStatement object, 20-2 OracleConnection class, 4-21 OracleConnection interface, 24-2 OracleConnection object, 20-2 OracleDatabaseMetaData class, A-9 OracleDataSource class, 8-2, 24-2 oracle.jdbc. package, 4-19 oracle.jdbc., Oracle JDBC extensions, 2-8 oracle.jdbc.OracleCallableStatement interface, oracle.jdbc.OracleConnection interface, 4-21 cancel() method, 4-21 commit() method, 4-21 getDefaultExecuteBatch() method, 4-21 getDefaultRowPrefetch() method, 4-21 getTransactionIsolation() method, B-4 setDefaultExecuteBatch() method, 4-21 setDefaultRowPrefetch() method, 4-22 setTransactionIsolation() method, B-4 oracle.jdbc.OraclePreparedStatement interface, oracle.jdbc.OracleResultSet, 11-5 oracle.jdbc.OracleResultSet interface, 4-23 oracle.jdbc.OracleResultSetMetaData interface, 11-14 using, 11-14 oracle.jdbc.OracleSql class, A-10 oracle.jdbc.OracleStatement, 11-5 oracle.jdbc.OracleStatement interface, 4-22 defineColumnType(), 4-22 getRowPrefetch() method, 4-22 setRowPrefetch() method, 4-22 oracle.jdbc.OracleTypes class, 4-23, 23-17 oracle.jdbc.pool package, 24-3 oracle.jdbc.xa package and subpackages, 28-5 OracleOCIConnection class, 24-2 OracleOCIConnectionPool class, 24-1, 24-2 OraclePreparedStatement interface, 4-22 getOraclePlsqlIndexTable() method, 4-29 getPlsqlIndexTable() method, 4-29 registerIndexTableOutParameter() method, setPlsqlIndexTable() method, 4-29 OraclePreparedStatement object, 20-2 OracleResultSet interface, 4-23 getXXX() methods, 11-9 OracleResultSetCache interface, 17-4 OracleResultSetMetaData interface, 4-23 OracleServerDriver class defaultConnection() method, 7-2 oracle.sql data type classes, 4-5 oracle.sql package data conversions, 11-3 described, 4-5 oracle.sql.ARRAY class, 16-1 and nested tables, 4-8 and VARRAYs, 4-8 getAutoBuffering() method, 16-4 methods for Java primitive types, 16-4 setAutoBuffering() method, 16-4

4-22

4-22

4-23,

4-29

setAutoIndexing() method, 16-4 oracle.sql.BFILE class, 4-8 oracle.sql.BLOB class, 4-8 oracle.sql.CHAR class getString() method, 4-13 getStringWithReplacement() method, 4-13 toString() method, 4-13 oracle.sql.CharacterSet class, 4-12 oracle.sql.CLOB class, 4-8 oracle.sql.data types support, 4-6 oracle.sql.DATE class, 4-9 oracle.sql.Datum array, 4-33 oracle.sql.Datum class, described, 4-5 oracle.sql.NUMBER class, 4-9 oracle.sql.ORAData interface, 13-14 oracle.sql.ORADataFactory interface, 13-14 OracleSql.parse() method, A-10 oracle.sql.RAW class, 4-9 oracle.sql.REF class, 4-8, 15-1 getBaseTypeName() method, 15-2 getValue() method, 15-3 setValue() method, 15-3 oracle.sql.ROWID class, 4-13 oracle.sql.STRUCT class, 4-7, 13-3 getAutoBuffering() method, 13-6 getDescriptor() method, 13-3 getJavaSQLConnection() method, 13-3 getOracleAttributes() method, 13-3 setAutoBuffering() method, 13-6 toJDBC() method, 13-3 oracle.sql.StructDescriptor class createDescriptor() method, 13-5 OracleStatement interface, 4-22 OracleTypes class, 4-23 OracleTypes class for typecodes, 4-23 OracleTypes.CURSOR variable, 4-16 OracleXAConnection class, 28-6 OracleXADataSource class, 28-6 OracleXAResource class, 28-7, 28-8 OracleXid class, 28-12 ORAData interface, 4-3 additional uses, 13-18 advantages, 13-8 Oracle object types, 13-1 reading data, 13-16 writing data, 13-17 orai18n.jar file, 19-2 othersDeletesAreVisible() method (database metadata), 17-17 othersInsertsAreVisible() method (database metadata), 17-18 othersUpdatesAreVisible() method (database metadata), 17-18 OUT parameter mode, 4-31, 4-32 outer joins, SQL92 syntax, A-10 ownDeletesAreVisible() method (database meta deta), 17-17 ownInsertsAreVisible() method (database meta data), 17-17

Index-7

ownUpdatesAreVisible() method (database meta data), 17-17

P
parameter modes IN, 4-29 IN OUT, 4-31 OUT, 4-31, 4-32 password, specifying, 2-8 PATH environment variable, specifying, 2-4 PDA, 18-8 performance enhancements, standard vs. Oracle, performance extensions defining column types, 23-17 TABLE_REMARKS reporting, 23-19 performance optimization, B-1 Personal Digital Assistant (PDA), 18-8 PL/SQL limit on BLOB size, 14-4 restrictions, D-2 stored procedures, 2-13 PL/SQL index-by tables, 4-28 mapping, 4-32 scalar data types, 4-29 PL/SQL types corresponding JDBC types, 4-29 limitations, A-11 PoolConfig() method, 24-5 populate() method, 18-7 positioning in result sets, 17-2 prefetching rows, 23-15 suggested default, 23-15 prepare a distributed transaction branch, 28-11 prepareCall(), 20-3 prepareCall() method, 20-6, 20-7 prepared statement passing BFILE locator, 14-16 passing LOB locators, 14-4 PreparedStatement object creating, 2-11 prepareStatement(), 20-3 prepareStatement() method, 20-6, 20-7 code example, 20-6 previous() method (result set), 17-10 put() method for Properties object, 8-10 for type maps, 13-9

3-2

Q
query, executing, 2-9

R
racgons, 26-4 RAW class, 4-9 read-only result set concurrency type, readSQL() method, 13-10, 13-11 implementing, 13-11 REF class, 4-8 Index-8 17-3

REF CURSORs, 4-15 materialized as result set objects, 4-15 refetching rows into a result set, 17-16, 17-18 refreshRow() method (result set), 17-16 registerIndexTableOutParameter() method, 4-29, 4-31 arguments int elemMaxLen, 4-31 int elemSqlType, 4-31 int maxLen, 4-31 int paramIndex, 4-31 code example, 4-31 registerOutParameter() method, 11-11 relative positioning in result sets, 17-2 remarksReporting flag, 23-15 Remote Method Invocation (RMI), 18-8 resource managers, 28-2 result set auto-commit mode, B-2 getting BFILE locators, 14-15 getting LOB locators, 14-3 metadata, 4-23 Oracle extensions, 11-5 using getOracleObject() method, 11-6 result set enhancemennts positioning result sets, 17-8 result set enhancements concurrency types, 17-3 downgrade rules, 17-7 fetch size, 17-15 limitations, 17-6 Oracle scrollability requirements, 17-3 Oracle updatability requirements, 17-4 positioning, 17-2 processing result sets, 17-10 refetching rows, 17-16, 17-18 result set types, 17-2 scrollability, 17-2 seeing external changes, 17-17 seeing internal changes, 17-17 sensitivity to database changes, 17-2 specifying scrollability, updatability, 17-5 summary of visibility of changes, 17-18 updatability, 17-2 updating result sets, 17-11 visibility vs. detection of external changes, 17-18 result set fetch size, 17-15 Result Set Holdability, 3-6 result set object closing, 2-10 result set types for scrollability and sensitivity, 17-2 result set, processing, 2-10 ResultSet class, 2-9 ResultSet() method, 16-4 Retrieval of Auto-Generated Keys, 3-4 return types for getXXX() methods, 11-9 getObject() method, 11-7 getOracleObject() method, 11-7 return values

casting, 11-10 RMI, 18-8 roll back a distributed transaction branch, 28-11 roll back changes to database, 2-11 row prefetching and data streams, 12-9 ROWID class CursorName methods, A-11 defined, 4-13 ROWID, use for result set updates, 17-4 RowSet events and event listeners, 18-3 overview, 18-1 properties, 18-2 traversing, 18-4 run-time connection load balancing, 22-1 enabling, 22-2 how it works, 22-1 Load Balancing Advisory, 22-2 overview, 22-1

S
savepoints transaction, 3-3 to ?? scalar functions, SQL92 syntax, A-9 Schema Naming, 4-4 scripts, authenticating users in, 9-21 scrollability in result sets, 17-2 scrollable result sets creating, 17-5 fetch direction, 17-10 implementation of scroll-sensitivity, 17-19 positioning, 17-8 processing backward/forward, 17-10 refetching rows, 17-16, 17-18 scroll-insensitive result sets, 17-2 scroll-sensitive result sets, 17-2 seeing external changes, 17-17 visibility vs. detection of external changes, 17-18 scroll-sensitive result sets limitations, 17-6 security authentication, 9-3 encryption, 9-6 integrity, 9-6 Oracle Advanced Security support, 9-1 SELECT statement to retrieve object references, 15-3 to select LOB locator, 14-10 sendBatch() method, 23-5, 23-7 sensitivity in result sets to database changes, 17-2 server-side internal driver connection to database, 7-1 server-side Thin driver, overview, 1-3 session context for KPRB driver, 7-4 setAsciiStream() method for writing CLOB data, 14-5, 14-6 setAutoBuffering() method

of the oracle.sql.ARRAY class, 16-4 of the oracle.sql.STRUCT class, 13-6 setAutoCommit() method, B-1 setAutoIndexing() method, 16-4 setBFILE() method, 14-16 setBinaryStream() method for writing BLOB data, 14-5 setBLOB() method, 14-4 setBytes() limitations, using streams to avoid, 12-9 setCharacterStream() method for writing CLOB data, 14-5, 14-6 setCLOB() method, 14-4 setCursorName() method, A-11 setDefaultExecuteBatch() method, 4-21, 23-4 setDefaultRowPrefetch() method, 4-22 setDisableStatementCaching() method, 20-6 setEscapeProcessing() method, A-7 setExecuteBatch() method, 23-4 setFetchSize() method, 17-15 setFixedCHAR() method, 11-13 setMaxFieldSize() method, 23-18 setNCLOB() method, 14-4 setNull(), 11-5 setNull() method, 11-11 setObejct() method, 11-11 setObject() method for BFILES, 14-16 for CustomDatum objects, 13-16 for object references, 15-4 for STRUCT objects, 13-6 to write object data, 13-18 setOracleObject() method, 11-11 for BFILES, 14-16 for BLOBs and CLOBs, 14-4 setORAData() method, 13-15, 13-18 setPlsqlIndexTable() method, 4-29 arguments int curLen, 4-30 int elemMaxLen, 4-30 int elemSqlType, 4-30 int maxLen, 4-30 int paramIndex, 4-30, 4-32 Object arrayData, 4-30 code example, 4-30 setPoolConfig() method, 24-4 setREF() method, 15-4 setRemarksReporting() method, 23-19 setResultSetCache() method, 17-5 setRowPrefetch() method, 4-22 setString() limitations, using streams to avoid, 12-9 setString() method to bind ROWIDs, 4-14 setTransactionIsolation() method, B-4 setValue() method, 15-3 setXXX() methods Oracle extended properties, 8-5 setXXX() methods, for specific data types, 11-11 signed applets, 5-3 Solaris shared libraries, 28-20

Index-9

specifiers database, 8-11 SQL data converting to Java data types, 11-3 types, constants for, 4-23 SQL engine relation to the KPRB driver, 7-1 SQL syntax (Oracle), A-7 SQL92 syntax, A-6 function call syntax, A-10 LIKE escape characters, A-9 outer joins, A-10 scalar functions, A-9 time and date literals, A-7 translating to SQL example, A-10 SQLData interface, 4-3 advantages, 13-8 described, 13-10 Oracle object types, 13-1 reading data from Oracle objects, 13-12 using with type map, 13-10 writing data from Oracle objects, 13-14 SQLInput interface, 13-10 described, 13-11 SQLInput streams, 13-11 SQLNET.ORA parameters for tracing, D-4 SQLOutput interface, 13-10 described, 13-11 SQLOutput streams, 13-12 SQLWarning class, limitations, A-12 SSL and LDAP, 8-13 start a distributed transaction branch, 28-8 statement caching explicit definition of, 20-3 null data, 20-8 implicit definition of, 20-2 Least Recently Used (LRU) scheme, 20-2 Statement object closing, 2-10 creating, 2-9 statement.cancel(), D-3 statements Oracle extensions, 11-5 stopping statement execution, D-3 stored procedures Java, 2-14 PL/SQL, 2-13 stream data, 12-1, 14-5 CHAR columns, 12-5 closing, 12-8 example, 12-3 external files, 12-6 LOBs, 12-6 LONG columns, 12-2 LONG RAW columns, 12-2

multiple columns, 12-7 precautions, 12-8 RAW columns, 12-5 row prefetching, 12-9 UPDATE/COMMIT statements, 14-6 use to avoid setBytes() and setString() limitations, 12-9 VARCHAR columns, 12-5 stream data column bypassing, 12-7 STRUCT class, 4-7 STRUCT descriptor, 13-5 STRUCT object, 4-7 attributes, 4-7 creating, 13-5 embedded object, 13-3 nested objects, 4-8 retrieving, 13-3 retrieving attributes as oracle.sql types, StructDescriptor object creating, 13-5 SYS.ANYDATA, 4-16 SYS.ANYTYPE, 4-16

13-4

T
TABLE_REMARKS columns, 23-15 TABLE_REMARKS reporting restrictions on, 23-19 TAF, definition of, 27-1 TCP/IP protocol, 8-12 testing for NULL values, 11-4 Thin driver applets, 5-2 LDAP over SSL, 8-13 overview, 1-2 server-side, overview, 1-3 time and date literals, SQL92 syntax, A-7 tnsEntry, 8-4, 28-20 toDatum() method applied to CustomDatum objects, 13-8, 13-15 called by setORAData() method, 13-18 toJDBC() method, 13-3 toJdbc() method, 4-6 toString() method, 4-13 trace facility, D-4 trace parameters client-side, D-4 server-side, D-6 transaction branch, 28-1 transaction branch ID component, 28-12 transaction context for KPRB driver, 7-4 transaction IDs (distributed transactions), 28-3 transaction managers, 28-2 transaction savepoints, 3-3 to ?? transactions switching between local and global, 28-4 to 28-5 Transparent Application Failover (TAF), definition

Index-10

of, 27-1 TTC error messages, listed, C-10 type map, 4-3, 11-6 adding entries, 13-9 and STRUCTs, 13-10 creating a new map, 13-10 used with arrays, 16-8 used with SQLData interface, 13-10 using with arrays, 16-12 type map (SQL to Java), 13-6 type mapping BigDecimal mapping, 13-32 JDBC mapping, 13-31 object JDBC mapping, 13-31 Oracle mapping, 13-31 type mappings JPublisher options, 13-31 type maps relationship to database connection, 7-3 TYPE_FORWARD_ONLY result sets, 17-5 TYPE_SCROLL_INSENSITIVE result sets, 17-5 TYPE_SCROLL_SENSITIVE result sets, 17-5 typecodes, Oracle extensions, 4-23

example, 23-12 executing the batch, 23-10 intermixing batched and non-batched, 23-13 overview, 23-8 update counts, 23-11 update counts upon error, 23-13 update conflicts in result sets, 17-14 update counts Oracle update batching, 23-6 standard update batching, 23-11 upon error (standard batching), 23-13 UPDATE in a result set, 17-12 updateRow() method (result set), 17-12 updatesAreDetected() method (database metadata), 17-18 updateXXX() methods (result set), 17-12, 17-13 updating result sets, 17-11 url, 8-4 URLs for KPRB driver, 7-2 userid, specifying, 2-8

W
WebRowSet, 18-10 WIDTH, parameter for APPLET tag, 5-8 window, scroll-sensitive result sets, 17-19 writeSQL() method, 13-10, 13-12 implementing, 13-11

U
unicode data, 4-10 updatability in result sets, 17-2 updatable result set concurrency type, 17-3 updatable result sets creating, 17-5 DELETE operations, 17-11 INSERT operations, 17-13 limitations, 17-6 refetching rows, 17-16, 17-18 seeing internal changes, 17-17 update conflicts, 17-14 UPDATE operations, 17-12 update batching overview, Oracle vs. standard model, 23-2 overview, statements supported, 23-2 update batching (Oracle model) batch value, checking, 23-5 batch value, overriding, 23-5 committing changes, 23-6 connection batch value, setting, 23-4 connection vs. statement batch value, 23-3 default batch value, 23-3 disable auto-commit, 23-3 example, 23-7 limitations and characteristics, 23-3 overview, 23-3 statement batch value, setting, 23-4 stream types not allowed, 23-3 update counts, 23-6 update batching (standard model) adding to batch, 23-9 clearing the batch, 23-10 committing changes, 23-10 error handling, 23-12

X
XA connection implementation, 28-6 connections (definition), 28-3 data source implementation, 28-6 data sources (definition), 28-2 definition, 28-1 error handling, 28-14 example of implementation, 28-15 exception classes, 28-13 Oracle optimizations, 28-15 Oracle transaction ID implementation, resource implementation, 28-7 resources (definition), 28-3 transaction ID interface, 28-12 XAException, 28-12 Xids, 28-12

28-12

Index-11

Index-12